SolidWorks Sucks...

John H,

I couldn't disagree with you more. It's ridiculous to say that a software took over the market because it has an easy to remember name. Being less expensive than ProE was a factor, but there still has to be a good product there to hold the customer for years. Kia does have 600,000 owners, but how many of them will buy a Kia as their next car? With SolidWorks, retension has been vital.

As far as the "as many flaws as strengths" statement. Well, the grass is always greener somewhere else. :) However, there's a reason ProE fell far behind and why Inventor doesn't dominant the market (which should've happened a couple years ago if what you are suggesting was true). It's because it is a great package for the price. And ya'no, it's even a better package for the price now more than ever.

Matt

Seems to me some user group experience might help. It might also help to have a seasoned user come in and review "how" you are trying to do what you are doing. With SW there is usually an easy way and a hard way. Many times the old ways are the hard ways as they come with a bunch of hidden assumptions that don't apply to solid modeling. However, even with all that SW can be slow as it really should let you do it your way and still perform well.

TOP

I think the MOST missed thing here was the fact he is using 2006 sp2.1 !!!!

That was a crap service pack and was slow as a dead snail.

Update to 2006 sp5.0, you will see a massive improvement and it seems to be rock steady, (for me, anyway)

One thing I do have a gripe about, are hole patterns, these really slow things down.

A good way to speed up designing, is to put commonly unchanged parts and assemblies, into the design library folders.

Here another reason for slowness that is quite often overlooked.

example Main assembly sub assembly a part a part a1

sub assembly b part b part b1

Open and change "part b" from the main assembly

If you do not open and re-save "part b", drawing for "part b", "sub assembly b", drawing for "sub assembly b" and finally re-open and save the main assembly and the drawing for the main assembly, this will slow everything down.

You can see this, when you go to check in the main assembly drawing, into Pdmworks or similar package.

The drawings will be opened and then saved, BEFORE, they will check-in.

Pete

I suppose you haven't heard of Beta and VHS then? Inventor has as many seats as SW. ProE doesn't because although it's technically superior it doesn't have the vendor network with local training and sales etc as much as SW and Autodesk, that goes a long way.

Retention of SW is because of market share. SW has reached a point where the market share drives the sales. It would be a completely different situation if 3D mech file formats were 100% cross compatible but that's impossible.

Instead of speaking in generalizations perhaps you could share with us specifics as to why ProE has fallen behind in market share? It's more complicated and more expensive for sure but it's definitel better.

That helps with selection but not always. The bigger frustation is the inability to change the thickness and/or hide/show etc. The hide/show problem has been around FOREVER.

Oh, how I miss the days of the drafting board. No file management, just throw it in a drawer. The smell of ammonia from the blue print machine. Electric erasers on Sepia copies. I'd go back in a heart beat.

good ol' bob z. ran into the inability to dimension a line about a week or so ago. it was the most frustrating thing for him at the time because it needed to be done and done right now. ya know? one of those pressure cooker scenarios. the line was just a line on the bottom of a plate. nothing special about it at all. bob z. is used to having a bit of an issue picking the edge of a circle, but this was a flat plate.

bob z. doesn't complain about speed. bob z. just finished a part that had 10,668 holes. now, that was a resource hog!!! it would've been cool if it had been a flat part, but it was a bell shaped detail, so each row of holes had to be it's own feature. (can't get into much more specifics, non-disclosure...)

other than that, swx rawks hard. it is just the little things, ya know? the little things. :~)>

bob z. is discussing this post with a co-worker right now. he does the big stuff - he has an assembly right now with over 8000 parts. oh yes.

bob z. p.s. kill the king. the king is dead. long-live the king.

You just need proper training. I was trained by Bob Zee. I have only been using SolidWorks for about 4 yrs. and have done several large machine designs completely in SolidWorks. The last machine had

10,000+ parts. I haven't had issues at all. Perhaps you could contact Bob Zee and get a little training on how to manage larger assemblies.

A little creating the drawings sure. Still better than Autocad so I don't complain

Thanks for the advice Pete!

I just installed 2007 SP0. I got to say It's still pretty slow, for the drawing part of it. I still can't get rid of the hatching either. I mean a simple cross section. The tools option is set at none and the auto hatching in the section view is checked of as well. I do not want to right click all the hatching individually and turn it off, which takes 51 seconds to do (yes I counted). Does anyone else has the same problem?

Cheers!

MDesign

Really? How the heck would you know that I need training or not. I know how to deal with large assemblies already, believe me I've been chasing the large assembly thing for years. I have machines that are feature intensive and although they have some subs this particular machine is mostly top level with many top level mates. If you had a machine with 10,000+ parts and had no problems then it was because you had it entirely broken into subassemblies with few top level mates and in lightweight mode with simple parts. My product structure does not allow for that.

Try most of the parts being top level, feature rich parts, weldments, sheet metal and so on combined with the fact that the assembly needs to be dynamic with moving hydraulic cylinders, splines and gears. It doesn't work well and it has nothing to do with my training level. Keep your assumptions to yourself.

Using Cylinders, gears and splines and then expecting them to be dynamic is amazing. You got me. The last machine I designed that was 10000+ components was actually comprised of 10000

1x1x1 blocks.

I guess instead of "chasing the large assembly issues" I have focused on finding ways to work with it instead of against it.

I have to agree with Bob Zee's acolyte here. What you're describing says to me that you're not taking advantage of the best methods for handling large assemblies. Stuff that VARs may not teach. It certainly seems that SW could be faster, but it sounds to me that you're not getting all the speed (or un-slow) out of SW that you could.

You mention a few things that I might recommend (not necessarily lightwieght) that your product structure does not allow. How is it that your product is not compatable with large assembly management techniques?

obviously the splines and gears aren't dynamic, that would never happen. The machine is dynamic in other ranges of motion driven by hydraulic cylinders.

speaking of 10000 blocks there is a benchmark you can use for large assemblies creating a thousand blocks with a hole through the middle and filleted edges. A colleague used this benchmark between SW 06 and ProE and ProE smoked SW. The company he worked for at the time was a manufacturer here that was using SW and had about 30 seats but have since switched to ProE.

What I'm referring to specifically is that I'm unable to use much in the way of subassemblies. The nature of the part requires that the assembly of parts be top level. The top level BOM needs to call the parts individually, not as subassemblies. A lack of subassemblies leads to an inordinate amount of top level mates. Add to this that many of these parts are resource hogs and the top assembly becomes a bear.

This is further complicated by the fact that most of are parts are weldments that are then machined. We want to use the weldment features adn show our fillets because the machine is tightly packed. We then call these parts in a second part where the machining drawing is created. We could use configs but this means one part file for two drawings which creates it's own problems with our document management. So we have a wldmt part called in a mach part that is then called in the top assy. These references seem to start to kill SW in terms of performance.

Of course the drawings are where the real pain is. The assemblies are slow but the speed of drawings and the difficulties in getting true high quality ANSI drawings is painful.

Why can't you use the "Show parts only" option on the BOM? With that all parts, even those in subassemblies, will be shown in a single BOM. Even if subassemblies did show, you can exclude them from the BOM using the BOM Contents option in the Properties of the BOM.

Am I missing something?

I do have some subassemblies too though. I could be wrong but I think if I "show parts only" then I'm showing all parts, no subassemblies even thought there are subs that I want shown as subs. Know what I mean?

I was just testing it out and I don't see any way to show the parts of a sub assembly in the BOM and show the subassemblies that I want to show as subs. Maybe there's a work around I'm not thinking of but I can't see how to do it. If I could it would be helpful, then I could reduce my top level part/mate count and speed things up.

Have you tried the check mark box under configuration properties that says "Don't show child components in BOM when used as a sub-assebmly"