Jump to content

B. Ludin

Community Members
  • Posts

    867
  • Joined

  • Last visited

Everything posted by B. Ludin

  1. FDM (Fused Deposition Modelling) parts are generally better than STL with respect to mechanical stability. However, the resolution of FDM is usually lower than that of STL (let alone microSTL). Beat
  2. B. Ludin

    Decals

    BTW, if you have Photoshop at hand, you can generate the desired image to use as a decal or texture quite easily. Just take a plain white background, use the "Add Noise" filter to generate some monochromatic gaussian noise, and use a wide "Motion Blur" filter to smear the noise horizontally. That should look quite convincing. Have a look at the attached scene. If you want it even more realistic, you could use the texture a a bump map. Cheers, Beat paper.zip
  3. B. Ludin

    Decals

    Just out of curiosity: why would you want to use a decal and not a texture? Beat
  4. Works fine with my GeForce2Go 32MB. Alas, the order dependent rendering makes it pretty much useless for me. Beat
  5. B. Ludin

    Laptops

    I have a Dell Inspiron 8000, 1GHz, 512MB RAM, 20GB HD, 32MB GeForce2Go, UXGA screen and I must say that I still like it a lot, although it's now 18 months old. The 1600x1200 screen is a real blessing, particularly for CAD applications and the graphics card does a decent job for the scenes I usually work with (typically 50-200 parts of various complexity). Based on that, I would can recommend the current Inspiron 8200 with the latest nVidia or ATI graphics card (although Dell's service over here in Switzerland isn't quite as impeccable). Cheers, Beat
  6. Sorry if that one gets posted twice, but it seems the first post didn't make it... ----- Thanks a lot, Maurizio This works like a charm (I just had to replace the slashes by backslashes in the path). How do you find out about things like that? The manual, and if so where? Source code? Or simply trial&error? Anyway, thanks again. Beat BTW, I would be interested in knowing the price for IronPRO.
  7. Thanks Cary - but sorry, I'm aware of how to drive a sizebox value by a parameter. But, as I described, I want to have it exactly the other way around, i.e. tie a parameter to a sizebox value, i.e. have it update automatically when I change the size of the object by dragging the sizebox handles (or any other action). In the case of extruded shapes it works, sort of, by adding an unlocked dimension to the 2-D cross-section. But this is only updated when I enter cross-section edit mode which is to tedious and can lead to unexpected results. Ideally, I would like to place an unlocked smartdimension measuring the distance between two faces of the shape (or whatever) and then use that to drive the locked smartdimension(s) which determine the position of the hole shape (or whatever). Unfortunately, unlocked 3-D dimensions don't appear as parameters, unlike their 2-D siblings, which actually constitutes an inconsistency in the parameterics implementation (take that as an enhancement request). Cheers, Beat
  8. Currently, the default format for callouts is something like 4.0x0.75. Is it possible to change the default (to something like M4) or at least to select all callouts automatically, so they can be formated together in one go? Beat
  9. Hi folks! I'm just starting to use parametrics to design re-usable parts. Works pretty well in general but there's something, apparently simple, that I can't find out: How can a parameter by driven by a sizebox dimension? For instance, I've got a hole that I always want to stay at 1/3 of the length of a block, no matter how I resize the latter (by dragging the sizebox or shape handles of the block intellishape). Can anybody give me a hint? TIA, Beat
  10. If, at some point, you want to reveal the non-rounded number you can just copy the rounded value and paste it back into the same place (ctrl-c-v) and it will appear at full precision. Cheers, Beat
  11. Just set up a user on your workstation which has the Novell clients disabled. That'll give you a direct comparison. Beat
  12. I'll throw in my IronMiles (provided I have any) for the plane-constrained triball, too. And - closer to the topic of the thread - for a modifier key that controls whether a intellishape droped onto a part becomes a member or a stand-alone part. These small enhancements would be the greatest mouse-click savers in my daily work (and it was actually a disappointment not to see them realized in IC5.2).
  13. id=quote> Hi Cary On my machine, it actually lists only one source - the one the last sheet is referring to, as far as I can tell. Cleary, the way it should be is that all the sources are listed and that you can change them individually. Forgive me my impertinence, but I'd still put that in the bug fix request section Cheers, Beat
  14. I've been able to reproduce at least part of your problem, i.e. when I have a drawing with two sheets using different sources and then change the source for one using crtl-shift-L, the other sheet changes to the same source (or in other words, changing the source acts globally). Once you're caught in that loop, there's no way out other than deleting the sheet. This is clearly a bug. Cheers, Beat
  15. I've had no problems yesterday using ftp://65.218.189.86/products/ironcad/5.2/IC52SP1.exe Also, for those, like me, who installed IC5.2 from the download and not the CD, you need to have the installation files in the same directory where you had them during the installation for the SP1 updater to work. Hope that helps, Beat
  16. Thanks, dlalonde, for having a look at the DXFs. In the meantime, I've spent many hours trying to nail down what exactly is happening. What I found is utterly confusing and to describe everything I've tried would take far too much time. Anyway, here are a few highlights: I've tried to produce identical DXFs on both machines by starting from the same .ics file and repeating exactly the same steps so that I could compare the files bytewise. Result: now the computer which was producing immaculate DXFs before has also started producing corrputed DXFs. Totally weird. It goes further than that. I can export a DXF, import it again and it looks perfect. I import the same file for a second time, now the dimensions are truncated to integers! This doesn't work just with the same file. I.e. I can export file A, import once (OK), import it twice (corrupted), then export file B, import file A (OK), import it again (corrupted). This is totally reproducible (at least a dozen times with several restarts inbetween). Now it gets really interesting: On both computers, I have a customized regional setting with the delimiter set to dot (I always had it like that) and an english keyboard setting. I also have an swiss-german keyboard setting which I only activate when writing letters in German. Now comes the interesting part: deleting the swiss-german keyboard remedied the problem. Re-installing it does not seem to have a negative effect (yet?). Here's a clumsy attempt to an explanation: At some point, one part of IC "discovered" the swiss-german keyboard and decided to use a comma delimiter when exporting a DXF despite the fact that the swiss-german keyboard was inactive and that the system was set to use dot as the delimiter. Seems like IC is still aware of this when importing a DXF immediately after an export, so the commas are interpreted correctly. However, the memory of this is erased as soon as the import process finishes, so a second import fails to recognize the comma delimiter. Probably some kind of strange interplay between WinXP and IC. The gist of it: IC developers: Have a look at how your DXF import/export routines handle language/keyboard/delimiter settings. Everybody else: Double check your DXFs if you have more than one keyboard setting. As I mentioned before, the problem here started out of the blue, after hundreds of successful DXF exports. It didn't coincide with the update to IC5.2, but I don't know if it is specific to that version and/or WinXP. Beat
  17. Ah, I'm glad that I get an reply after all. We definitely must solve the problem. We just had to trash parts for nearly $3000 which were cut incorrectly because of this problem/bug (there goes a second seat of IC...) so I'm a rather upset as you might imagine. Anyway, here's some more information. The only system difference between the two computers is that the one on which IC exports fine has a US-english keyboard while the other one has a swiss-german keyboard. That would support your hypothesis of a dot/comma problem. HOWEVER - the delimiter is set to dot (and I never touch these settings) on both machines. Moreover, I've just been able to export a screwed-up DXF again and I checked to make sure the delimiter was set to dot at the moment before and while IC was running. I'm uploading a corrupted DXF for you to analyze. I'm not familiar enough with the DXF format to figure out what's wrong. Beat The original .ics 10356A_OK.DXF 10356A_SCREWED.DXF 10356A.ics 10356A.DXF
  18. I have IC installed on my home and office computers. When I export 2-D DXFs from a drawing it works perfectly on one machine whereas the resolution is reduced to whole units (mm, in my case) on the other (so a hole with 4.34mm diam. becomes 4.00mm, etc.). Does anybody know where the setting is hidden that has this effect? I can't seem to find it... TIA, Beat
  19. IronCAD is reading my e-mail??!? Bummer! Should I fear that it'll start actig up when I write bad things about it? (Sorry, Chris, couldn't resist)
  20. The "missing symbol syndrome" with PDF export from IC doesn't seem to be a problem of missing fonts, since it occurs also when the PDF is opened on the same machine. Printing to PDF with Docucom or pdf995 works correctly (although the former screws up character spacing sometimes) in the same environment. So for the time being, my favorite remains pdf995. It's cheap ($10 or even if you don't mind pop-up advertisement) although the interface for the interesting special functions (e.g. page combination) is rather cumbersome. Beat
  21. Tip/Bug Report: When you save a part/assembly from a scene with several configurations, always remember to switch back to the default configuration first, otherwise the part may get corrupted (around 10% chance, I'd say from experience). This is particularly important if you have "link to scene" checked, because in that case you won't have an original to go back to. Beat
  22. Dave - thanks for the hint. Tom - Rhino works with 3D digitizers. But where are the photogrammetry functions hidden? BTW, if anybody else is interested, here are some of the programs I found: PhotoModeler Pro ShapeCatcher ImageModeler 3D Builder Pro I have tested PhotoModeler Pro in some detail and achieved accuracy in the range of 1:2000 with an uncalibrated digital camera. Their website claims that 1:20000 is possible with a properly calibrated high quality digital camera! Just for those who are interested. Beat
  23. Hi folks This question is not directly related to IC, but I thought one of you CAD cracks might be able to help me. We build custom-designed incubators for microscopes. Our customers often send us pictures of their existing setups from which we would like to extracvt some 3-D information. Do you know any software that would help us to do that? In general, the observer location will not be known but would have to be inferred from known absolute and relative properties of objects (typically lines) in the picture(s). Ideally, the software should also be able to deduce distortion parameters of the camera optics. TIA, Beat
  24. This is not new to 5.2. I've also seen this behavior in 5.0 and I think in 4.2 also (I've only worked for a short time with that version). It's really worth restarting IC from time to time to keep the performance up. Annoying, for sure, but a phenomenon quite common to software of such complexity (which shouldn't be taken as an excuse not to fix it). Beat
×
×
  • Create New...