How about a simple time clock that measures time spent modeling..... >
>
> > Awright...
> >
> > I'm grumpy now.
> >
> > I have spent a considerable amount of time (like a year) developing
> > something that I think a whole stack of SolidWorks users would love. I
> > succeeded. Unfortunately, my release date co-incides almost exactly with
> > the release of SW2004.
> >
> > This would be fine, except that SW2004 (as I have recently discovered) > just
> > so happens to impliment about 80% of the functionality that I had
> > implimented. When you get it, check out what they call weldments... > you'll
> > love it.
> >
> > In the process of developing this, I have managed to construct many useful
> > components (data structures, user interfaces and the like), and learnt a
> > small truckload about the SW API (which - full credit - is fantastic). > >
> > Which brings me to the reson for this post. I would be very interested in
> > developing something else now - as I have learned all this and would like
> to
> > make use of the knowledge - but no longer have a need to satisfy. > >
> > If you feel inclined, please let me know what you would like to see that
> > would improve your day/week/month (what I had would take a month back to a
> > day) - at this stage, I don't care how big or small, I am after the germ > of
> > an idea.
> >
> > Cheers to all.
> > Woylie Engineering
> >
> >
> >
>
>
How about an application that maximizes (or minimizes) a given model dimension to the least or greatest value that works? Like a fillet, I want the max value fillet that will fit.
How about a program that makes custom property creation and management easier.
How about a lettering add-in that allows you to wrap text around a surface or curved solid. It would also allow you great control over the fonts, line spacing and kerning files. If there was a way to read the 50 or so not truetype fonts I have that would also be great
I'd like that to, but for the opposite reason. Parasolid will tolerate a fillet value that's larger that what will actually fit mathematically. This causes all kinds of problems downstream when you try to machine a mold cavity. The oversize fillets end up with surface errors that drive CAM systems (and their users) crazy. Sometimes the files don't even translate completely.
If the utillity gave you the real mathematical maximum, without going into tolerant mode, I'd install it on all our I.D.'s machines. It would save me allot of time working around these problems, and editing.
I'm working right now on the last function needed to extrude them normal to the surface...
Which brings to the main problem : users don't know the existing add-ins, and they don't buy because they expect SolidWorks to implement that function in the next release. Based on our experience, it takes SolidWorks about 3 years to integrate a feature that exists in an add-in, and they won't warn you unless you specifically ask a friendly insider... Moreover, VARs aren't interested in add-ins (too few, too low price) except if they help winning a SW market over a competitor... And you have no way to contact users directly at an acceptable cost (mailing list) In other words, I'd say the "Solution Partner" program doesn't work.
Tracey, I have several projects that need one more good programmer. I can't pay a salary, but if you are interested in a partnership, call me!
I bet that anyone working on this had/has/will have the same bad surprise as Tracey very soon... Someone at SolidWorks told me one day that I was "running in front of the bus" with my add-ins. I keep this in mind all the time : "you have to run faster than the bus", and use narrow roads where it won't go...
How about a program that can split part configurtions into seperate part files. Example. The existing part file contains configuations for a casting (A) and 4 machined versions (B, C, D & E). I would like this program to split the file into 5 part files, part A being the casting, parts B, C, D & E being the machined version but having part A as the base part. To create a base part can be an option in this program, maybe even link back to there respective drawings! If you use a PDM system (Smarteam) you will know why, and yes we would pay for this app.
How about a true macro recorder. One that memorizes menu picks and plays them back without requiring the CAD user to edit the macro in a programming language. This would work much like the macro recorder in ProE.
Being able to assign hot keys that use two alpha characters would also enhance the usefulness of macros tremendously.
Surprising... do you have any link/info on this ? Or a very simple example so that I can figure out this problem ?
Ouch, it would mean to use an alternate geometrical representation than the SW/Parasolid... Or do you think it would be enough to find the minimum distance between the edge to fillet and any other (non intersecting) edge ? Or should the face curvature also be takein anto account ?
I would have to agree that there are a lot of addins out there that we don't know about, but since I don't know about them then here are a couple of ideas. I heard here lately a request for wire harness design software and the existing SW partner is no more. I have done wire harness design using the methods previously, and while adequate, I think automating effectively could be beneficial. Also, I beleive there is tolerance analysis software available, but if not, maybe this would be good as well. Anyway, whatever you decide to take on next, good luck and keep us informed. Oh yeah, feel free to ask us for specific functionality because people who influence the design are always more likely to accept the design.
It's a face curvature problem. When the faces get "crowded" they degenerate. This usually happens at a point (singularity ?). These degenerate surfaces apparently have tiny (wrinkles ?) in them, and the surface normals shoot off in weird directions. I've even seen them reversed in small areas, (haven't seen anything this bad ffor awhile though). You can imagine what the machining routines do when they encounter them. There are work arounds in my CAM system, but they don't always give optimal results (especially with really small details).
If there was a utility that would, at least, flag degenerate areas, it would help allot.
Create an offset surface offset by the tool radius. Offset that surface by a small amount. Is there an intersection between the two offset surfaces? If so you have a problem ....
You can (perhaps) also look at surface normal reversals in the first offset surface but ....
You do need surfaces .
On some systems writing a quick program to do this sort of thing is easy .... and also a good way to select the minimum size ball endmill that can cut ALL of the surface(s) selected. It's a little smaller than the offest is when you get intersections of the offset surfaces.
Somewhere else in this thread, Phillppe noted that 'Someone at SolidWorks told me one day that I was "running in front of the bus" with (-) add-ins'. That is of course the rub - if the macro/add-in is really useful, SW will take the inspiration and add it into their software (unless of course it is spell-checking for drawings and notes, which the folks in Concord still can't navigate a solution for!)
They are a multi-million dollar corporation, and you are not (I presume)... you might be able to take advantage of that fact to stay ahead of 'the bus'.
For instance, your add-in could take advantage of licensing things that would cost SW too dearly because of their volume (i.e. I suspect that licensing a dictionary for a spell checker is holding them back, because, frankly, the folks at SW cares too much to have let that obvious enhancement go this long without a fix).
Or you can take advantage of their corporate mind set and exploit a blind spot. Here's an example - Every user encounters rebuild errors in the first few minutes of using the software, but no VAR I know of has written a course for how to deal with them, and I suspect that SW, for marketing reasons, would not want to invest in lots of error handling tools (Inventor or PTC could point out that SW has so many 'errors' they had to write a suite of tools to handle them) Error editing macros would be easy to write, and would be sale-able for a long time to every level of SW user - PLUS, SW corp is not likely to get off of the new-feature bandwagon and get on the-fix-what's-there-and-simplify-and-damn-the-marketing-fallout bandwagon.
Tools I would like to see:
- The 'oh sh**' button - I just made a change to a model that blows up lots of features. I can't close out of the model because, darn it, I wanted the five edits that led to the eruption. However, I can't figure out how to fix the issues because I don't know what the bleeding features were supposed to look like (a fillet shows no edges - where was it supposed to go?) Right now, I go grab the copy off the hardrive, copy it to a new folder, rename it, then open it in a separate window of SW so I can roll through the 'before' and 'after' trees side by side and compare features. Wouldn't it be great to have a simple 'oh sh**' button that automates that? - RMB a feature and see what it looked like before I screwed it up?
- The 'artistic sketch edit' button. I modify a lot of sketches based off of what the sketch looked like before - I want it to 'bulge out a little bit more', etc. I currently copy the sketch before editing, then paste it on the some plane, then roll it back before the original sketch so I can use it as an underlay for my changes. I'd pay a buck or two to get a button to automate that. I'd really love it if it also did 3-D curves, model edges, etc (open a 3D sketch, convert the original curve into the 3D sketch, delete all relations in the 3-D sketch,then move it up the tree). This button would also be a handy way to save design history in you part, so you can go back to/compare to an earlier version!
-Undo a change to sketch. This is a stupid, stupid, stupid, stupid oversight on the part of SW corp that you could fix in your add-in.
-Flip plane normal. Lots of sketches flip when edited to a new plane, and you can't always get them back to look right (if the sketch has external relations, you can't use modify sketch to flip or move, and Paul Kellners terrific edit-to-a-perpendicular-plane-then-edit-back trick sometimes causes the sketch to freak out)
-Dangle child relationships when deleting parent features. I have a bombproof manual technique for this (sabotage the parent feature, discussed before on this group) but I would so much rather a simple button that I could press that would delete a feature without killing its kids.
There are all sorts of other editing tools that come to mind, but its getting late. All I can say is take advantage of any blindspot SW might have that the user base would not have. Real time deformation of surfaces - well, duh. Error handling tools, spell checks, automating adding prefixes or suffixes to file names, etc - maybe those would be under the radar of the folks in Cocord, insuring some ROI.
Have you heard something? I would love it if the 'bus' got there, but assumed that it wouldn't because these basic issues have been overlooked for so many years. BTW - When I've talked to SW developers at SolidWorks world and trade shows they seemed too conscientious to just let stuff like this go - there is always some background hurdle, like Windows limitations, timing, etc. holding them back
Many of the suggestions you make are not just running in front of the bus, but very much stepping out in front of a bus type suggestions... suicidally so.
You idea of an "oh S**t" button is great... but...
How on earth can you impliment it? I mean, what you are saying (If I read you right) is you would like an undo button on steroids? You would like the undo button to not only read you mind as to how far back to undo, but to figure out which bits of the last indeterminate number of edits should be undone?
OR... are you saying that you would like to have a record of all the edits that ever happened to the model, and let the user search back thru the model? This one is fairly easily implimented - but - it will be a killer on resources, and to be neat, will do what I have on good authority as being "Nasty things to the SolidWorks file"... not something I would like as a user.
I really do like the idea in principle though, so if I am misunderstanding, please fill me in!
Please see my other response to your original post also.
No I haven't. I meant it is impossible from a commercial point of view for partners to correct bugs in SW:
1) users won't buy the product because they expect SW to do it in the next SP or major release, and they feel they paid for it in the subscription already.
2) advertising would be based on criticizing the SW product, which is forbidden by the partner agreement we signed.
3) implementing low-level features such as Undo Sketch through API requires much, MUCH more work than for SW. (Remember the API basically let you automate only things that you could do by hand...) FYI, I fixed the Flip Plane problem in the Pattern3D feature of SolidSketch
formatting link
), which lets you create the same extrusion on many different planes. This way I can advertize for the added value feature (3D patterns) instead of the bug patch, but you can "3D pattern" a sketch on a single plane...
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.