B. Ludin

Thanks for the offer, I'll get back to you. But what do you mean by "PRO/E IGES files by default are too inaccurate to be of much use"? Firstly, the file in question is a STEP file not IGES, and secondly, we have used both PRO/E STEP and IGES files in the past and never noticed any inaccuracies, just mistakes like false orientations and missing surfaces. Or is the latter what you meant? BTW, one intersting observation is that IC reports bad entities when importing to ACIS but not to Parasolid. That's a bit puzzling, isn't it? Beat

I'm trying to import a 65MB ProE-generated STEP file (header below) on a 1GB RAM machine and get a out-of-memory message from the SmartHeap Library. The task manager suggests that there's still some 150MB of physical memory available and the page file is not even at 50% of its maximum. I've tried reducing the smoothness and import to both ACIS and Parasolid, but to no avail. Any ideas (apart from getting more RAM) how to get the problem solved? Is this a problem of heap size and is there a way to allocate more memory to the heap? And no, going back to the provider of the file to ask them to give us another format or break the model down into smaller parts is virtually impossible. We have received the file as a courtesy and it took our contact in that company more than 8 weeks of persistent nagging, until their oh-so-busy CAD guys were finally willing (or sufficiently unnerved, rather) to spend the 15 minutes to export the model. So this file is what we have to work with. TIA, Beat ISO-10303-21; HEADER; FILE_DESCRIPTION((''),'2;1'); FILE_NAME('000000-1299-411_01-205_ASM','2004-09-06T',('zjspe'),('<company name deleted>'), 'PRO/ENGINEER BY PARAMETRIC TECHNOLOGY CORPORATION, 2003090', 'PRO/ENGINEER BY PARAMETRIC TECHNOLOGY CORPORATION, 2003090',''); FILE_SCHEMA(('CONFIG_CONTROL_DESIGN')); ENDSEC; DATA;

They do indeed, thanks. Beat

He Carlito - ned übermüetig wärde gäu! BTW, folks, Carlo had a really and rightfully succesful art presentation (called "the big loop") last weekend on the piazza at the west entrance of Solothurn's old town. His metal sculptures were complemented by dynamic graphics, sound/music and large projections on the surrounding builidngs. Splendid! If anything's worth coverage on IC's web site, this is. Chris, you should get Carlo to provide some pictures so you can show off with the works of your customers. Cheers, Beat

Hey bkline How about getting an IBM T221? With its ginormous pixel number (3840x2400, that's the equivalent of 12 XGA or 5 UXGA monitors!) you wil probably have enough screen real estate. Just make sure you get a REALLY powerful graph card to feed this baby. Just an idea. PS: IBM had that thing for a fair while already but they don't advertise it because demand exceeds production anyway. And no, I have absolutely no commercial interest in IBM. Cheers, Beat

Tom - I can't seem to download the file using your link (I guess the Irondudes have removed it already). Can you re-post it the regular way, I'm interested in it, too (I've tried to reproduce what you described but it din't work). Beat a mere sphere, but at least it rhymes... ;-)

I get an error saying "Sorry, but you do not have permission to use this feature. If you are not logged in, you may do so using the form below if available." when I try to access a member profile, the top ten posters, etc. Why is this so and does anybody else have the same problem? Beat

I'm pretty sure this is not currently possible in IC. Othwise, it should also be possible to make cutouts from unfolded sheetmetal and then refold it (an enhancement that has been requested a couple of times in the past). Beat

Yes. Unfortunately, this is not true for the 2-D construction environment (consider that an ER). Beat

665MB

Tom - most of the curved surfaces seem to be of the UV mesh type or could at least be modelled as such. But the model itself is impossible, i.e. there are surfaces which will never combine into a solid (THE big danger with surface-based modeling) and there's even a couple of lines which don't end in a vertex so they will never form a surface (the big danger with wireframe-based modeling?). Anyway. without knowing the intent of the creator, I think it will be difficult to clean the model up. Beat

Lovely!

Let me guess... you've created these in Poser? Great software if you want to populate your scenes. I can only recommend it. Cheers, Beat

Marvellous! I'm going to trash my Inspiron8500 on Monday and get myself an iPAQ. With a suitable holder and a handle bar pointing device (left handle controls X-axis, right handle controls Y-axis) I should be finally able to keep CADding while riding my bike! Beat

I'd like to put in my vote for an SDK (software development kit) together with API tutorials. But I think some very basic programming knowledge should be a pre-requisite. I wouldn't like to see the IC development team reinventing the wheel by writing basic programming tutorials. But a number of examples and comprehensive reference would be highly welcomed. Beat

I don't (ab)use the current functionality either. However, if you change it the way you describe, how would you handle associated views (sections, etc.), which usually depend on the orientation of the standard view? Beat

Of course. I should have thought about this myself. Thanks, Floris. Beat

I have a similar problem that I always wanted ot report but never got to. When I export assemblies as parasolid and import them into our CAM software (VisualMill), some parts are often mis-oriented and/or -positioned. The workaround is to export as ACIS, import, and then export the re-imported assembly as parasolid. I'm not sure if this is really IC's fault because the parasolid exports always look good when I look at them with HOOPS 3D, so it might be VisualMill's fault (or incomplete implementation on either or both sides). IC's support: it would be nice if you could look into this together with Mecsoft since you have a collaboration anyway (and maybe get them to provide a native IC importer?). Cheers, Beat So I wonder, I you re-import your

Cary - What speaks against having both, i.e. why don't you give the user the choice whether a Smartdimension should act as a sequential or a simultaneous constraint? Beat PS: I agree with the others, if the parametric features were better implemented and, just as important, better documented, they would probably be used much more often.

I voted both, but should add that it's 90% triball. Beat

I have no problems drilling down to subtracting shapes. I can also select the shape from the browser and then select it from the scene, if necessary. Beat

Corey- pitch = diameter * pi / sin(angle) The vertical helix crossection width w is much more difficult. I can't figure it out just like that. But a little trial and error will get you there. Actually the value that I've used w = width / sin(angle) = 0.5 is inaccurate. The value should be around 0.568. mmccall (whatever your real name is) - Yes, it's a loft you have in mind and most likely it won't work (accurately) in this case, or at least it will be very difficult to adjust the tangecy. But I think a 2-guided sweep might work. A solid sweep would be best. Cheers, Beat

I think you can do it with the helix tools directly. The attached part should show you how it's done, assuming you know your geometry (i.e. how to calculate the pitch from the diameter and the angle and projected crossection of the helix). Cheers, Beat Helix.ics

Excellent, thanks! Now I can finally assign undo&redo to my recently acquired spacemouse. Beat

Convert you part from Parasolid to ACIS and try again. For some reason, Parasolid runs into memory problems very quickly when you increase the surface smoothness. Cheers, Beat