Jump to content

P Retondo

Member
  • Posts

    1,914
  • Joined

  • Last visited

Posts posted by P Retondo

  1. Garry, since your screen is a field of pixels that is pretty coarse, compared to most printer output, you are going to see that jaggedness no matter what (they call it "aliasing"). Whenever your line crosses from one row of pixels to the next, there has to be a jump. The reason you are seeing it in an iso view is that you have angled lines that cross the rows of pixels. The more "shallow" the angle, the more noticable the jump. If you look at printer output with a magnifying glass, it will be jagged, too, but on a much finer grain.

    [ 02-28-2003, 08:19 PM: Message edited by: P Retondo ]

  2. If I attempt to clip a polygon with another polygon under certain circumstances, the first polygon is deleted instead of being clipped. I'm not sure how general this problem is, but I have at least one instance of it. In my example, the clipping polygon will divide the clipped polygon into two shapes. When the point of "division" is a vertex, the clipped polygon disappears. If I move the clipping polygon to either side of the vertex, I get the expected clip results. Pasting the shapes into a fresh file, the problem recurs without fail.

    This is yet another bug that has to do with objects having coincident boundaries. It seems to be a common problem. The functions have to be more robust than this!

    On the bright side, at least the program didn't crash!

    [ 03-02-2003, 12:56 PM: Message edited by: P Retondo ]

  3. One of the reasons it's so convenient to use the wall styles tool is that you can enter the thickness of the wall elements. The way the wall tool dialog box works, by contrast, is that you have to calculate the distance of the cavity line from the center of the wall. If you have 5/8" gyp bd on the inside, and 1/2" ply plus 3/4" siding on the outside, try figuring out where the cavity lines are relative the center of the wall. Not that it can't be done, but it would be much easier if the cavity lines were placed relative to the outside of the wall instead of relative to the center.

    [ 02-27-2003, 08:55 PM: Message edited by: P Retondo ]

  4. If you select the option to have interior and exterior trim of different widths, the 3D wall void is still cut to the outline of the larger trim size. This shows up as a gap in the wall in 3D when viewed from the narrow trim side ("Window" insertion tool). I note that all symbols/PIOs cut openings in walls at their extreme boundaries, in both 2D and 3D. In 2D, the symbol creates a likeness of the wall to fill in the gap between the edge of trim and the frame. In 3D, the fill is not created, causing the problem noted above.

    VW 10.1

    WinXP

    Is it outside the realm of possibility to have the wall objects cut at the door or window frame, instead of at the edge of the trim? If it is possible, this would take care of the problem, plus a few other annoying glitches - such as the facts that 1) often, in 2D, there is a line across the wall at one side of the window or door at the outside edge of the trim, and 2) when my wall cavites are not symmetrical I have to go though a series of flips to get the faked in walls to align their cavities correctly.

    [ 02-27-2003, 02:39 PM: Message edited by: P Retondo ]

  5. quote:

    Originally posted by jan15:

    The math for ellipses is pretty tough.

    NNA: here's a pretty good reason to consider including a way to divide a curved line into segments, as someone has requested recently. The math is tough, unless a computer can do it for us.

    BTW, I wonder if your programmers have ever used the C++ library that MatLab provides. Makes this kind of thing relatively easy.

    [ 02-27-2003, 01:55 AM: Message edited by: P Retondo ]

  6. This is normal behavior, but if you have a lot of saved views/sheets, it can be tedious to update them all. I have suggested elsewhere that a utility to make this easier would be great to have. In the meantime, I have discovered that if you edit the sheet setups via the RESOURCES dialog box, it can go a lot faster. Open the resources for your document, then open the saved sheets. You will see list of all sheets/views. Right click on each of them to edit. The list persists, so you can go down the list one by one and not lose track of where you have made changes and where you haven't, plus you don't have to reopen the edit dialog box for each sheet you want to change.

  7. To answer some complaints/suggestions heard in this discussion, I think it would be important to modify the "saved sheets" to include:

    Remembering the selected view setting for every visible layer. This will get rid of having to "Align layer views" when you go to a sheet and some layer happens to have been left in top or side view.

    Allowing different page sizes and other settings to be incorporated in to the Save View settings. Currently, it is useful that the page location can be remembered, but extending this to other plot-related parameters would be great.

  8. Draw a rectangle, and draw a line inside it so that one side of the line lies on a side of the rectangle. Now attempt to join the other side of the line to the rectangle using the Connect/combine tool in Single Object Connect mode. It will not connect. Now resize the line so that its end no longer lies on the rectangle. Now it will connect.

  9. zaklee, have you tried layer links? This allows you to plot any layer at any scale. As for whether the entities are "full scale," there is no difference between ACAD and VW on this point. Everything is full scale in VW, but VW skips the step you have to take in ACAD and allows you to "draw on a sheet" at a predetermined scale. But don't let that fool you. You can reset the scale on that sheet, or put layers into another sheet at another scale using links. The only thing ACAD has which VW doesn't is the ability to crop or mask links easily, so that you can insert only a portion of a layer via a layerlink.

    zaklee, from this and other comments you are making, I don't think you have given VW a fair chance. You need to explore the VW equivalents to your ACAD techniques. Once you do, you will never miss the kinds of arcane complexity that jan15 is talking about.

  10. I've posted some of the problems I've had with the pocket door (VW 10.0.1, .vso files dated 2-11-03). In addition to those, the pocket door in 3D open mode constructs a door that is wonky. Stiles and panels are in the wrong locations.

    Try looking at a pocket door that is 90 degrees open. It sticks out of the pocket, and has no end stile. A panel door that is < 90 degrees open has the panels incorrectly constructed.

    My wish is to see this door fixed!

    [ 02-25-2003, 02:05 PM: Message edited by: P Retondo ]

  11. Charles, you have said that you do not experience the problem I describe. What are the parameters of the system you are testing on? Can you confirm that what I describe is a bug using VW 10.0.1 on WinXP? Or do you believe there is some other source for the problem? If so, what is your theory?

  12. Hi Charles, yes, that is exactly how I have set the default class for my PIO door tool. The objects are assigned the correct class. The class is set to a solid fill (white), with attributes set to "use at creation." Yet the object is unfilled (wireframe in rendered 3D). I have to select it, and modify the fill attribute in the attribute dialog box.

    [ 02-25-2003, 02:22 AM: Message edited by: P Retondo ]

  13. B. Howell, what is probably happening, as RA is suggesting, is that although the object is set to a certain class, which is active, it contains within it a bunch of primitive objects which are members of a different class. Hence, with the "active class only" setting, you are looking at an empty "container".

  14. When I set the class for the "Door" PIO insertion tool, the object acquires the given class automatically. However, it does not acquire the attributes for that class automatically, even though the class has been set to "use at creation." I have to manually modify the attribute for fill every time I insert a door. I assume the same behavior is true for the "window" tool. This behavior does not occur with any of the "2" version tools. Am I missing something, or is this a glitch?

    VW 10.0.1

    .vso's dated 2-11-03

    Win XP

    [ 02-25-2003, 02:16 AM: Message edited by: P Retondo ]

  15. The new pocket door PIO is much, much better than it used to be. The pocket draws correctly, and all of the object attributes seem to work well. I have the .vso version dated 2-11-03, and it still has two significant glitches in the 2D department:

    1) The wall outline is drawn with a very thin lineweight. The pocket outline is okay, it's the outside wall line that is incorrect.

    2) Also, the walls of the pocket should have the same fill color (and other attibutes) as the wall!

    Is there any workaround for these problems, or do I just have to create some walls and fill overlays to make this look right?

  16. quote:

    Originally posted by Katerina:

    I wish most of the program's foundation commands had a 'redrawall' procedure at the end of the script.

    -as needed

    Katerina, doesn't "z" "z" redraw all objects? I don't do any vectorscript programming, but can't you simply invoke those commands? I think that calling a redraw procedure at the end of every command in the core engine might slow down the program very considerably, so I would recommend caution on this.

  17. When I move a layer up or down in the layers list, I can lose track of where that layer is. The list should autoscroll so that as I am moving a layer up I always see two layers above its current location; and when I move a layer down, I should always see two layers below.

    I do appreciate that we now have autoscrolling, it's just that I need to see what layers are near the layer I'm moving. In the past, we had to guess where in the list the layer was. It would also be nifty if by clicking and holding on the "down" or "up" button, we could continuously move the layer, similar to the way the nudge tool works.

    [ 02-23-2003, 05:03 PM: Message edited by: P Retondo ]

×
×
  • Create New...