Subject:
|
Re: LPub Rotation Step angle woes
|
Newsgroups:
|
lugnet.cad.dev.mac
|
Date:
|
Thu, 6 Nov 2008 17:40:57 GMT
|
Viewed:
|
11815 times
|
| |
|
|
In lugnet.cad.dev.mac, Allen Smith wrote:
| |
In lugnet.cad.dev.mac, Jim DeVona wrote:
| |
In lugnet.cad.dev.mac, Allen Smith wrote:
| |
I’m working on rotation steps, and I’m not seeing LPub do what I expect it
to do.
I have the following file:
1 2 0 0 0 1 0 0 0 1 0 0 0 1 3865.dat
0 ROTSTEP 0 0 0 ABS
1 4 -150 -24 70 1 0 0 0 1 0 0 0 1 3005.dat
0 ROTSTEP 0 0 0 REL
|
|
|
|
|
Please note that instead of “0 ROTSTEP 0 0 0 REL”, the preferred meta command is
“0 ROTSTEP END”.
| |
| |
| |
That should give me:
- A step with a front view
- A step with the default 3D angle.
|
|
snip
| |
| |
Am I completely missing something here? My output is not oriented the same
way the pictures Jim posted in his tutorial were.
I am using LPub 4.0.0.1 with LDView 3.2.
Allen
|
Hi, Allen.
Here’s what I see when I open your example in LPub 4.0.0.1:
|
Alas. That is what I’m supposed to see, not what I actually do see. Something
is convinced that the 0,0,0 rotation is actually equal to the default 3D
angle, which causes requests for the real default 3D angle to be
over-rotated.
|
There are 4 kinds of ROTSTEPS: REL, ABS, ADD, END
For REL, you add the angles from the ROTSTEP meta to the default viewing angle
to get the final viewing angle. When you have a REL, ADD, or ABS rotation from
previous steps, REL tosses out the affect of the previous ROTSTEPs.
0 0 0 REL # should not change the viewing angle
For ABS, you discard the default viewing angle and use the value provided in the
meta as the final viewing angle. When you have REL, ADD or ABS rotation from
the previous steps, ABS tosses out the affect of the previous ROTSTEPs.
0 0 0 ABS # forces and end on view of the model
For ADD, you add the values from the meta to the “current” viewing angle. This
is like REL, but instead of adding the meta and the default view angle, you add
the meta to the “current” viewing angle (as possibly defined by previous ROTSTEP
metas). When you have REL, ADD, or ABS rotation from previous steps, ADD
modifies the affect of those previous ROTSTEPs.
For END, the current view angle is defined as the default view angle.
| |
| |
I’m not sure why the difference, but I’ll list some possible factors.
- Side effects of other LPub options, e.g. to hide PLI
|
|
No.
| |
| |
- LDView version (I use a “4.0 Pre-Alpha” version from SourceForge/Travis)
|
|
No. LPub handles ROTSTEPs for the renderers.
| |
| |
- Assembly image cache - try clearing it (from the Tools menu) and clicking the Redraw button
|
|
Possibly. The most up to date caching mechanism takes rotation into account, so
the unrotated version of the step would have a different name the the rotated
name.
| |
Well, it’s definitely not #1 or #3. Kevin or Travis, do you have any idea
what is going on here?
|
I hope the above explanation helps.
Are you having trouble within BrickSmith, or using files created in BrickSmith
with LPub?
If this is the second case, when I get back from my work trip, I’ll look into
it.
| |
| |
A confession: I never use ROTSTEP ABS (I’m just in the habit of using REL),
so I’m not very familiar with its behavior. I rotate things relative to the
default view.
I probably misuse ROTSTEP, too (but I put it to good misuse). I usually put
ROTSTEPs immediately after STEP commands (at the top of the file, for the
first step) and before the parts I intend to rotate. So, my version of the
file might look like this:
0 ROTSTEP 0 0 0 ABS
1 2 0 0 0 1 0 0 0 1 0 0 0 1 3865.dat
0 STEP
0 ROTSTEP 0 0 0 REL
1 4 -150 -24 70 1 0 0 0 1 0 0 0 1 3005.dat
|
|
Or, in the case of LPub multi steps, I put the ROTSTEP immediately after the
multi step begin statement (which goes immediately after multi step end,
which goes immediately after the STEP...).
Hopefully someone else can STEP in and clarify. I’m afraid I use a lot of
voodoo to insert these commands.
Jim
|
As Kevin wrote, the form I posted is the correct one. This was a fairly
recent epiphany on my part, which is why Bricksmith currently leads you
astray in where it is putting step commands. 0 STEP/ROTSTEP is supposed to
come after the parts in the step, not before. And ROTSTEP replaces STEP. I
intend to fix this behavior in the next release of Bricksmith.
Allen
|
Please let me know if I can be more help.
Kevin
|
|
Message has 2 Replies:
 | | Re: LPub Rotation Step angle woes
|
| (...) I know. I used the form I did to confirm that the default viewing angle was mathematically messed up (and for dramatic emphasis). (...) I've got them, now it's time to test them. (...) Yes, actually, it was LDView that was confused. Travis' (...) (16 years ago, 7-Nov-08, to lugnet.cad.dev.mac, FTX)
| | | Re: LPub Rotation Step angle woes
|
| (...) Well, Allen said that the LDView 4 test build fixed the problem. At a guess, LPub using -cg on the command line (to set the camera angle to 0 so that all the rotation can be in the model itself?) and the QT version isn't handling that like it (...) (16 years ago, 7-Nov-08, to lugnet.cad.dev.mac, FTX)
|
Message is in Reply To:
| | Re: LPub Rotation Step angle woes
|
| (...) snip (...) Alas. That is what I'm supposed to see, not what I actually do see. Something is convinced that the 0,0,0 rotation is actually equal to the default 3D angle, which causes requests for the real default 3D angle to be over-rotated. (...) (16 years ago, 6-Nov-08, to lugnet.cad.dev.mac, FTX)
|
12 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
|
|
|
|
