HELP HELP i have to talk sense in people with no CAD knowledge

Such an interesting story - and we all thought we had it rough. :-)

WT

A most interesting story! We are seriously considering going the other direction. We've been running SW since 98+, but working with fairly complex surfaces is driving us nuts. It's taking us about three times as long to build our parts as it should, if only SW worked the way it is advertised. We are looking at Unigraphics, Catia and Pro/E.

With reference to Pro/E, you mentioned:

Dealing with the PTC sales force has to rank right up there with my least favorite things to do in the whole world.

I wouldn't count them out yet. They are still a force to be reckoned with.

Gee, I've felt the same about SW since about SW2000. They keep adding these nice capabilities that don't quite work in real life.

What was your benchmark? How did Pro/E fail?

Well, SW has been a success here, but the last two projects have taken much too long to develop, with way too many hours on nights and weekends.

Pro/E was never easy to learn, but it was always fairly powerful. I would guess that with training and constant use that you might be faster with Pro/E if it only took you a little over twice as long to build the part with no training.

They will suck your blood no matter what you do. It's part of the corporate culture.

Jerry Steiger Tripod Data Systems "take the garbage out, dear"

My reply point is simple:

The easiest thing in the world for a manger to do to make sure his purchase "works" is to "Overbuy" so "I don't make a mistake". With that in mind, suggest that if they want to never make a mistake with a product that won't do what you need, then buy Catia. If they want to be efficient in software price, training & time, suggest that they find something better than SolidWorks at the same price.

Bo

just curious Jerry...what type of product are those surfaces for?

Look at the Ranger and the Recon on the TDS website:

Ranger is the one on the right on the top page and was done with SW98+. Industrial design (hah!) by me, no surfaces, and looks like it. Recon is the one on the left, industrial design by real designers in Pro/E, surfaces duplicated (more or less) by us in SW2001+ The conversion into parts took much longer than expected. The new product has industrial design done in Alias and duplicating those surfaces and turning them into real parts has also taken much longer than planned. We're much better at making the surfaces, but are plagued with long read and write times, crashes and having to build stuff over and over again as SW flakes out.

Jerry Steiger Tripod Data Systems "take the garbage out, dear"

Jerry i have a lot of experience with complex surfaces. Surfacing of coca-cola bottles etc.. Catia will be more powerfull then pro-e for surfacing it will even offer you the possibility to model complex 3d artwork and engraving. But the price :-( will be loads higher then pro-e) because you will need additional surfacing options. But what to you get in the place is always interesting extra features with new releases.

You have also another very interesting product for surfacing

this is a familly of software from delcam going from surfacing to complex engraving, 3d 5 axis milling and reverse engineering well worth looking at it. Very good helpdesk This product i love because it is so fast to learn and the most powerfull surfacer i know. One guy keeps 5 high speed milling machines on the boil on his own. A coca-cola bottle takes 1.5h to model and 15 minutes to program.3d programming has been extensivly automated. Machines are running 50% and 30percent less polishing time faster then when we do the programs with pro-e because it allows us to use the optimal milling strategies. surfacing is also half of pro-e time. in the mothercompany one guy make models and programs for 5 high speed milling machines. One of them is a real rocket running at 40000 rpm and hitting feeds of 120000mm minute for finishing. The guy is averaging 2 to 3 moulds a day. In the uk with pro-e they need 2 guys to run 3 high speed milling machines and they are even struggling even with the machines running slower. Even with this argument managment didn't want to allow the uk plant to switch softwares. But also pro-e users can be a pain to convince that there are some other softwares on the market. I guess they like the power they have in there hands. Because managment will not give the sack to quickly because they know they are up with a six month trainingto replace the guy for complex surfacing and machining with pro-e.

The only hick up the system is not parametric only to a certain level. But like you see it does'nt slow the programmer down. Because a lot of the time when you want to change something in a parametric model with complex surfacing some features are failing and then al the fun starts specially with pro-e. You can say think it over how to built it before you start but a lot of the time when it is really complex you have to try different ways to get there and it starts to be a mess. I was only doing the effort if when it was a familly of bottles with a range of different volumes. Then it was worth it to put the energy in to it.

Around the mould cavity parting line there is a groove who has to be parallel to the parting line. If you switched cavity this groove had to regenerate automaticly to be parallel with the mould cavity. The problem is that the curve that you use to make offset varies from one bottle to another. When you switch cavity's it doesn't regenerate because number of entities are not the same in the seed curve.

Well they solved it by making a spline from the curve. But pro-e does only make combined splines if the profile is tangent so no corners are allowed above the 5 degrees. They didn't tell us and they received some samples with curves. So afterwards if i wouldn't have tried myself we would have been stuck with it and would have to program something to solve it. I know vb not c++. Must say at ptc they have mighty well trained sales people.

sounds like you could/should move to something more compatible with the design people you work in with rather than try to replicate shapes in SW that are obviously more complex than the program can really handle. SW2005 looks like a step in the right direction but as far as continuities and nurbs go its still in the mid-range. No doubt your Ranger is a good design but it shows how organics have moved into ID and it makes it look somewhat dated. this type of difficulty with soft shapes is what concerns me about SW even with the coming enhancements. its a competitive world...can I afford to produce B grade shapes? and spend extra time doing it?

Hehe - We used to refer to them as the "Cad Mafia" . . .

(their-mouths-full-of-cotton-marlon-brando-esque) "So, ya gonna buy, er waht? You want I gotta go see ya boss? We're rubbin ya current CAD system out . . . See . . . (I quitely think - please guys don't use the term Deliverables again or I will hurl)"

I love that stuff. I'm a pretty laid back sort and I eventually had to tell the pro-e guy to "piss off and don't call me ever again" (or something pretty close). I felt bad but he stopped calling, so all in all it was a good thing in the end.

:)

SMA

That's the way we figure it too. We're going to be looking at Pro/E, Catia and Unigraphics as well as ShapeWorks and GeometryWorks.

SW2005

One of us has the beta running at home. He hasn't been too impressed so far. I guess I shouldn't talk about what he said in this forum.

Ranger is a good design

You're either very kind or blind as a bat. It's ugly.

We don't need to do automotive type surfaces, but we do want to make more attractive products. SW just isn't cutting it right now. I hate the thought of moving to another system, but we can't afford to waste our precious engineering man-hours. We've probably lost something like three engineering man-months on this project. Call it $100 an hour, 40 hours a week (really much more), and 12 weeks. That's about $50K down the tubes.

Jerry Steiger Tripod Data Systems "take the garbage out, dear"

The price, of course, makes us nervous. What makes me even more nervous is the possibility that after paying the big price we still run into time-wasting road blocks. That's why hearing from people like you is very important.

I'm very hesitant to look at software that doesn't integrate nicely with a solid modeler. We still have lots of detail to put in that would be a pain to do in surfaces.

This is where I get very nervous. We could have just taken the Alias surfaces from the ID firm, but what happens when we decide that we need to add a mm here or there? (That actually happened on the Recon project and is bound to happen in the future.) Even if we build the surfaces ourselves with another piece of software and then integrate them with our solid modeler, I think the wasted time factor might become a problem again. Or am I just being too nervous?

Because a lot of

I'm sorry to hear that Pro/E also ran into these same problems. That is what has been driving us nuts with SolidWorks. You make a nice shape, possibly after a lot of trial and error to get it to work. Then you make some seemingly simple change that doesn't seem like it would affect the original shape. Suddenly the shape is broken and you spend half a day trying to get it working again. Make a couple of simple changes and spend several days trying to get rid of all of the cherries.

They aren't people. They are sales droids. When I was using Pro/E (almost six years ago, now) I met some support people that I really liked. The programmed selling machines drove me crazy. The thought of putting myself in their hands again sends chills up my spine.

Jerry Steiger Tripod Data Systems "take the garbage out, dear"

Try using Rhino3d with Solidworks. Then you have the best of both worlds