Jump to content

jan15

Member
  • Posts

    1,090
  • Joined

  • Last visited

Posts posted by jan15

  1. I support all 3 wishes.

    1. It would be nice if objects outside the group being edited could be greyed when they're visible. I always have the outside objects visible, but I often make the group red temporarily while editing it. It's easier when you can see what the group consists of.

    2. I use a macro utility similar to Quickeys, and a keyboard with 24 function keys, to get all the menu commands I commonly use mapped to unshifted keys. It seems to me VW would be very slow without it. It should be possible to do that in the workspace editor. I don't see why it shouldn't be possible to assign unshifted number and even letter keys to menu commands as well.

    3. As you said, assigning attributes by keypress is a quick way to work but cumbersome to set up even with a macro utility. I finally gave up and just use the eyedropper tool, which is sometimes a good alternative and sometimes not. It would be nice if there were some way the workspace editor could set that up.

  2. The spacing of hatch lines can be in Page units (relative to the printed page, such as .05" spacing for shading) or World units (relative to the real-world object being portrayed, such as 8/3" spacing for brick joints).

    Editable by clicking on the word "Hatch" at the bottom of the pull-down list of hatches in the Attributes palette.

    Those units always apply to the hatch when first created. For the associative hatches created by the Attributes palette, the units adjust when you change layer scale or re-size the boundary object, to be the same as if you had applied the hatch after the scaling or re-sizing. Non-associative hatches created by the pull-down menu command are really just groups of lines, and they act like any other lines when you change scale or re-size.

  3. Sorry about that. I forgot. It's hard to believe, and hence to remember, that they don't make a Mac version of the viewer. I guess with their shoestring budget they can't afford any frills.

    What about MacDwiff? at http://www.macdwf.com/

    And I remember reading that CadViewer is supposed to read DWF's on a Mac, and generally on any kind of machine that has Java installed. Not true? http://www.cadviewer.com

    Can you post the DXF file so we can open it and see what's going on?

  4. And DWF is a viewing-only format ? Autocad can't open it either. There's a viewer for it, a free download from Autodesk.

    But it sounds like the DXF file they sent you is an older version, so you shouldn't be having problems with it. If using the viewer doesn't sort it out, tell us more about what you get when you import the DXF file, or post the file.

  5. In Windows, the filename extension tells the operating system what program to use when you double-click on a file. You could rename a text file "notes.mcd" and Windows would tell VW to try to open it.

    MacOS uses a different method, and ignores the extension.

  6. I don't use Constrain-to-Grid, but I tried it one time and it seemed pretty easy and foolproof as long as it was the only constraint turned on, and as long as Units was set to the same number of decimal places as the snap grid. In fact, with those settings, I think it would be easy to draw without typing in any numbers, just dragging and then clicking when the right numbers appear.

  7. What exactly do you mean by "change the size?" How are you making that change?

    VW is very precise, but there are ways of working that don't take advantage of it. For example, drawing or drag-stretching by watching the data bar numbers till the right one appears sometimes produces sizes slightly different than what you think you're getting. The numbers generated by the data bar are an approximation of the cursor location, affected by your Units settings, snap grid settings, and constraints settings.

    But any numbers you type in the data bar are treated as precise, not approximations. If you type 300 and 150 in the data bar while dragging, or in the OIP, or use a 0.5 factor in the Scale Objects tool, you should get a rectangle that's 300.000000000 x 150.000000000.

    Sizes of objects always have a very high precision, not affected by the Units settings. That only affects display. And the snap grid setting only sets the interval to snap to. If you snap to 300.004, or if you type 299.999998, that's the exact size the object will always be, though in both cases it will display as 300.00 whenever Units is set to 2 decimal places.

  8. There's also a stretch technique not similar to Autocad's: select the objects to be stretched, hit the Group command, then grab any of the 8 handles on the group and stretch it to a snap point, or by a distance typed on the keypad.

  9. Yes, there've been many problems with offsetting polylines that include curves. I always avoid it and find some other way to accomplish the task.

    In the case you cited, I would offset the 6" square first (drawing a new 6" square between the non-filleted opposite corners if necessary), and then fillet the corner of the resulting 5" square to a 2.5" radius, using the third mode of the fillet tool. There's no need to de-compose, but if you do the arc is correct.

    The problems with offsetting curved polylines have been much discussed here. There should be a pop-up warning when you try to do it.

  10. ... wishlist the option to import SketchUP Layers as Classes rather than as Design Layers.
    You'd think that would be the standard way, or even the only way, since Sketchup Layers are similar to VW Classes and don't have the extra features of VW Layers.
  11. Newly created objects are assigned to the active class and layer, which are shown in two boxes side by side in the data bar at the top of the screen. Objects pasted in are pasted onto the active layer, but they keep whatever class they had before. You can pull either box down and make a different class or layer active.

    Those boxes always show the active class, regardless of whether any objects are selected. They don't show the class and layer of selected objects, as is the case with similar boxes in Sketchup and Autocad. The class and layer of selected objects can only be read in the Object Info palette.

    Since you used hyphens in your class names, they'll appear in the pull-down list as if they were called "intelligent" and "generic" and as if they were in a folder called "LX."

  12. Sorry no one answered sooner. I'm not sure I can help. Can you post the DWG file? That might help us to see why there are differences. Is the black image something they sent you, or something you got in your own DWG viewer or editor? If the former, did they get that when they opened your DWG file, or when they inserted yours as a block into one of their own?

    There shouldn't be any substitution for Arial or Arial Narrow truetype font, since Autocad has been able to use truetype for many years now, and everyone has those two (unless your client removed Arial Narrow from their system Fonts folder for some reason). The only other possiblity I can think of is that they might have another font called Arial Narrow that their Autocad prefers over the TTF that you're using.

    Autocad complicates the font issue with text styles. A text entity doesn't have a font as one of its properties, it has a Style. Each Style is associated with a font and a width factor and some other things, plus the text entity can have font overrides that are assigned to some or all of the text in the editing window. So if they inserted your file into one of their own that already had a style called Arial Narrow, and if that style was associated with Arial regular or Swiss 721 truetype font or one of the gothic shx fonts, or had a large width factor... But that doesn't seem likely. If Autocad doesn't recognize a font when it opens a dwg file it just substitutes simplex.shx and doesn't say anything about it, but in your sample it doesn't look like that's what's happening.

    I've never had any problem exporting symbols or groups to DXF. I'm not sure I understood what you said about exchanging a block for a symbol. The standard thing in Autocad is that if you insert a block from another file into a file that already has a block with that name, then what you get is the block as it's defined in the current file. If you want to re-define a block based on one in another file, you have to insert the foreign block under a different name and explode it, then select it all and use it to re-define the domestic block.

    Any Group should be exported as a Block. If that block is then pasted into another DWG file, the visibility of objects in it would be based on what class you put them in and the visibility of AC "layers" in the new file.

  13. 1:8 is the scale we normally call 1 1/2" = 1' - 0", a common scale for molding details. It may be that they were drawn 1/8 actual size so they would print at 1 1/2" scale without having to use viewports. That method used to be standard practice in Autocad, which doesn't have layer scale. I know of people who still use it to put drawings at two different scales on the same sheet, and I've seen exactly what you described in manufacturer's drawings.

    Except for the units issue that Pat mentioned (and none of those would give you a 1:8 reduction), geometry always comes in at the size it was in the DXF file. Layer scale doesn't affect that; it only affects the relationship between virtual geometry and virtual print-out in the VW file. If it's as I suspect, the only solution is what you're doing now, re-sizing with the Scale Objects command. You shouldn't have to group them in order to do that.

  14. I think pierpaciugo means what the US version calls Sheet Layers. The best word for that in another language might be one that indicates a presentation or representation of the design on paper.

    And I think this discussion is based on exporting Classes as DXF Layers, which is the best way to go because Autocad's "layers" are similar to VW classes and don't have some of the features of VW layers (such as scale and layering). In this mode, VW Classes become AC Layers, and VW Layers are not differentiated in the DXF file.

    If you use the other mode, exporting Layers as DXF Layers, you may be disappointed because the layers can't take their scale or their layering with them. Autocad layers don't have those properties. They're all going to be at the same scale, with no distinction of what's in front and what's in back. They'll be differentiated only in the way that VW classes are.

    In that preferred mode of Classes as DXF Layers, it sometimes helps to export only one design layer at a time (or at least limit it to design layers that are all at one particular scale) because in Autocad a file can only have one modelspace, which is its nearest equivalent to design layers, and modelspace can't be scaled like design layers can. So anything on a design layer has to be exported full size to modelspace. All the design layers have to be exported to the same modelspace, with no differentiation, and all at 1:1 scale.

    That includes any text on a design layer. The text has to be exported at the size it would be if the design layer were scaled 1:1. For example, 12 point text on a design layer at 1/4" scale (1:48) is exported about 5" high in 1:1 modelspace, so that it will be 0.1" high when seen through a 1:48 viewport. That's how Autocad handles text in modelspace.

    You can put dimensions and text on a design layer if you want, no problem. VW will make the letters 5" high, or whatever is appropriate, in the DXF file, and the Autocad user will think 5" high letters are the most normal thing in the world. If you want to distinguish dimensions and text from the geometry on that layer, put them in separate classes (dimensions are automatically put in the Dimensions class anyway). Then they'll be on separate Autocad layers.

    If you want real Autocad compatibility, you either have to give up using the advanced features that Autocad doesn't have, or else spoon-feed it several files as a simplified version of your VectorWorks file. And switch to using VW classes rather than layers for anything you want to become AC layers.

  15. That doesn't sound strange. The viewer shouldn't care about read-only status, since it's only able to read. The editor should offer to open a copy of the file under a new name. Does it do that?

    I don't know what Apple xServe is, but the only cases where I've seen files become read-only on their own have been files on CD's. Under Windows 2000, those files remain read-only even when they're copied onto a hard drive, and the user has to un-click read-only in the properties window. It can be done to all selected at once. But you've probably already done that, since you say you confirmed that they're read-only. Is there still a problem, or are you just wondering how it happened?

  16. Petri, you're twice blessed, for not having to use imperial feet and inches, and for not having to use imperial Autocad.

    3 feet six and a quarter inches is written 3'-6 1/4". Base 10, base 12, and base 4, 8, 16, 32, or possibly 64 numbers all in one linear measurement. But that's a walk in the park compared to the pointless complexity of Autocad.

    Here in the "homeland" (now defended by a Department of Homeland Security, to distinguish it from the imperial security provided by the Department of "Defense"), scale is an issue.

  17. Everything is 1:1 in the sense that the user always enters the actual dimensions of an object. But layer scale makes a translation between the virtual model and the virtual sheet of paper, just as a wood or plastic strip with bevelled edges makes that translation for a drawing on real paper.

    What makes this relevant is that Autocad doesn't have layer scale. The translation has to be done by the user. Viewports help the user to do that, but some AC users still make the translation in modelspace, by entering incorrect dimensions for objects so that one drawing will print at a different scale than another. For example, the user might draw walls 8" thick in a 1/8" scale plan, but draw the same wall 64" thick in a detail to be printed on the same sheet, so that the detail will be scaled at 1" scale.

    Could that be the cause of the "scale wander" Steve? Sometimes it's done by scaled block insertion. AC Blocks, similar to Symbols otherwise, can have their insertions scaled up or down. So, following the example above, the user can draw the detail at actual size, then block the whole detail and insert it at 8 times its original size as a 1" scale drawing on an otherwise 1/8" scale sheet.

  18. If you paste-in-place the corresponding geometry from the same file, then the viewports would show the same thing they showed in that file, wouldn't they? I've done that in another CAD program. It's handy for cases where an older project has a lot of the same details you need for the new one.

×
×
  • Create New...