Subject:
|
Re: mouse model manipulation
|
Newsgroups:
|
lugnet.cad.dev.mac
|
Date:
|
Tue, 11 Jun 2002 02:09:44 GMT
|
Viewed:
|
1983 times
|
| |
|
|
In lugnet.cad.dev.mac, Jim DeVona writes:
> Well, I looked at the pdn and pan values, but they never seemed to get
> crazy or anything (although I'm not familiar with the inner workings of
> OpenGL so I'm not entirely sure what constitutes fishy). Their values stay
> relatively close together.
>
> In regards to the glutWarpPointer possibility, the model will continue to
> spin even after the mouse leaves the window, so I'm not sure if that's it.
Do you mean cursor doesn't disappear when you drag the mouse with the
button pressed? It's supposed to. That way you can't see it warp back
to the center of the window when it strays too far. Or does the cursor
reappear when you leave the window? It shouldn't do that either unless
the Mac has a really bad pointer focus policy.
> One thing I've realized on closer inspection is that the rotation works
> better in orthographic mode than in perspective mode (I'm working in ledit
> mode here)... although the direction of rotation still does not seem to
> correspond closely with the mouse movement, the model is drawn in a more
> proper orientation... it seems as if the smaller the mouse movement, the
> more accurate the final position of the drawn model (as compared to the
> wireframe during rotation), while large mouse movements result in greater
> discrepancies. In perspective mode it seems much less predictable.
That sounds a bit more normal. Is it just because the redraw is too
slow? Try the using the l3 parser by passing -l3 on the command line.
It refreshes faster because it doesn't reparse the files everytime it
has to draw them. It also calculates bounding boxes for the parts,
which also speeds up the default spin mode redraws quite a bit.
> You know, even if its not that glutWarpPointer function, I would agree
> that the problem is probably due to some Mac input glitch, not a problem
> with the ldglite code (especially if mouse rotation works well enough on
> other platforms). In any case, I think it will be less of an issue when
> the keyboard commands we've discussed are implemented.
>
> Its not really that serious of a bug, mouse rotation is certainly usable,
> just not very intuitive or effecient. But still it'd be nice to iron it
> out.
Ok I implemented the keyboard spin commands, and modified the mouse
spin calculations to be done in screen coords. That should help.
I don't know what I was thinking, spinning using perspective view
coordinates...
Which sounds better? Spin the model 180 degrees by moving the mouse
half the window width (or height). Or spin 180 degrees by moving the
mouse say 200 pixels?
Don
|
|
Message has 1 Reply:
 | | Re: mouse model manipulation
|
| (...) The cursor does not disappear when the you drag the mouse with button pressed. However, neither does it "warp" back to the center of the window when it leaves the window - the cursor just moves as expected. (Although since I don't know OpenGL (...) (22 years ago, 11-Jun-02, to lugnet.cad.dev.mac)
|
Message is in Reply To:
| | Re: mouse model manipulation
|
| (...) Well, I looked at the pdn and pan values, but they never seemed to get crazy or anything (although I'm not familiar with the inner workings of OpenGL so I'm not entirely sure what constitutes fishy). Their values stay relatively close (...) (22 years ago, 11-Jun-02, to lugnet.cad.dev.mac)
|
66 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
