PDA

View Full Version : Why Won't This Work?



Tony3d
03-01-2013, 09:34 AM
Can someone tell me why on God's green earth, that this won't shrink to this wire insolation? If it's because their hollowed out for the conductors, then Heatshrink is pretty useless for cables. Which is what I would mainly use i for.

bazsa73
03-01-2013, 09:42 AM
XswampyX did it for you!
http://www.youtube.com/watch?v=dB8NxvZqeGY

Tony3d
03-01-2013, 10:08 AM
Yes, I saw that. This is really quite simple, and I see no reason why it won't work? Two tubes twisted, one shrink. I don't get it. Just another feature that isn't quite finished yet.

Lewis
03-01-2013, 10:19 AM
what happens when you shrink it to 100% ? Can you send model so we cna check/try ?

Tony3d
03-01-2013, 10:27 AM
Yes, On phone with client right now. Will do that in a minute. I did try 100% with no luck. Has to be the hallow tubs. Must need a solid shape.

OnlineRender
03-01-2013, 10:34 AM
i feel it works better with numeric panel open

Tony3d
03-01-2013, 10:40 AM
I have the numeric panel open.

Tony3d
03-01-2013, 10:43 AM
Here are the models.

Tony3d
03-01-2013, 10:44 AM
Here are the models.

XswampyX
03-01-2013, 10:54 AM
Work here with my model?

http://i465.photobucket.com/albums/rr16/xXswampyXx/HShrink2_zps5e1086cf.jpg

You can't have any gaps between the cables. Use point normal move to fatten them up. then use heat shrink on the fat copy to get the effect. Then use thicken on the heat shrink object to get the heatshrink back to touch the original mesh.

Will try your model.

XswampyX
03-01-2013, 11:01 AM
I've had a look.....

You have too many polys in your cable? if you run reduce poly + (twice) then point normal move @ 2mm. It works fine. :D

Lewis
03-01-2013, 12:07 PM
Well it really don't work on this mesh, it's realyl dense but hey it still should WORK so id' say you found a BUG.

So my suggestion is (beside workarounds you already got ;)) is to Delete all 0-point polys (you have 520 of those and that's "junk" geometry), triangulate 4 end Ngons (yust so you exclude possibility they saing it's bad geo) and resave mesh and send that to Fogbugz and let NT deal wiht this BUG.

Or If you wish I can do that (fogbugz report) wiht this mesh.

cheers

jwiede
03-01-2013, 02:04 PM
If LW is going to be so intolerant of "junk geometry", then maybe it shouldn't make so much in the first place.

Lewis
03-01-2013, 02:10 PM
If LW is going to be so intolerant of "junk geometry", then maybe it shouldn't make so much in the first place.

I Agree on that (not making junk geometry). I'm askign for Junk geometry removal/kill (at least as option in prefs for automatic deletion) for years now but nothign happeneed and I think untill people realize that 1-point polys and 2-poitn polys are simple JUNK and just LW hacks (instead proper features) 'coz LWs inability to render points and splines we will probably constantly have to deal with that junk :(. I hate when i need to select/dele it hundred times a da yin stats panel (I'm nto exxgerating 'coz when modleign subD sand welding stuff/optimizing Lw constantly creates 2-point polys and then TAB SubPatches complains about it (CCs are even worse they make creases on mesh where 2-point poly sjunk is leftower) bla bla bla.....)

On the subject this mesh don't work even when cleaned of all junk so it must be BUG, Reported to NT few minutes ago so let's hope the will fix it soon.

jeric_synergy
03-01-2013, 02:21 PM
Obviously you don't HAVE to go via the stats panel for 1 and 2: that's a simple script, eh? (Not that I feel qualified to double-guess Lewis.)

Lewis
03-01-2013, 02:28 PM
Obviously you don't HAVE to go via the stats panel for 1 and 2: that's a simple script, eh? (Not that I feel qualified to double-guess Lewis.)

Well i have one scrip to kill 1-point/2-point polys Matt made for me long ago but it's problematic in some modes and it happens to delete too much sometime (sometime kills trangles also or selecte dpolys) so i'ts problematic and risky to use :). I'd just be happy wiht option for WELD/Weld Average called "don't creat 2-point polys" 0coz those two gives most junk geometry in my case and i use it very often :). Funny thing is that weld edges and weld edges loop don't create that junk geometry :).

jeric_synergy
03-01-2013, 02:31 PM
Well i have one scrip to kill 1-point/2-point polys Matt made for me long ago but it's problematic in some modes and it happens to delete too much sometime (sometime kills trangles also or selecte dpolys) so i'ts problematic and risky to use :). I'd just be happy wiht option for WELD/Weld Average called "don't creat 2-point polys" 0coz those two gives most junk geometry in my case and i use it very often :).
Huh! Considering all I do manually is hit "1 point polys" and DELETE, I wonder what the script is doing??? :stumped:

Lewis
03-01-2013, 02:38 PM
I dont' know, I'm not programmer, it has specific set of rules to be followed in special mode and you must be sure you dropped any selections before hittign tool/shortcut so it's kinda slow workflow then and it screwed me few times in past on dense meshes (i didn't notice it deleted some tris untill much later when i hit TAB) I don't use it often anymore :). I'm still thankfull that Matt took time to create it (and I'm sure now he would create even better one 'coz he is programming more and more, not just fancy interface drawing ;)).

jeric_synergy
03-01-2013, 03:36 PM
I dont' know, I'm not programmer, it has specific set of rules to be followed in special mode and you must be sure you dropped any selections before hittign tool/shortcut so it's kinda slow workflow then and it screwed me few times in past on dense meshes (i didn't notice it deleted some tris untill much later when i hit TAB) I don't use it often anymore :). I'm still thankfull that Matt took time to create it (and I'm sure now he would create even better one 'coz he is programming more and more, not just fancy interface drawing ;)).

What'd be nice (for this: and generally???) would be if we could set functions to be executed on system calls, sorta "Everytime you save, do this BEFORE/do this AFTER" kinda thing.

Of course, we could just usurp the SAVE function entirely and put it within a script. We'd 'think' we were just Saving, when in reality we'd be executing a number of scripts.

Lewis
03-01-2013, 03:43 PM
Nahhh, I need it during modeling not just at save, SubPatches/TAB contantly nag/complain if you have 1-point/2-point polys you always get error messahe "only polys wiht 3/4 points can be....."

ConjureBunny
03-01-2013, 03:47 PM
112075

It seemed to work here.

Or is this wrong?

-Chilton

XswampyX
03-01-2013, 04:38 PM
Ooooops going 'awf topic a bit. :D

Here's the tricky lwo after a couple of reduce polys + and the heat shrink applied.

112076

ConjureBunny did you mess around with the object or did it work 'out of the box' ? When I tried at first nothing, I repeat NOTHING happened with the heatshrink tool?

Thanks!

Spinland
03-01-2013, 04:49 PM
Worked for me as is when I changed normal mode to closest points. Left some stretched stuff along the seams but at least it did work.

Tony3d
03-01-2013, 08:22 PM
Finally got something I'm happy with!

ConjureBunny
03-01-2013, 09:05 PM
ConjureBunny did you mess around with the object or did it work 'out of the box' ? When I tried at first nothing, I repeat NOTHING happened with the heatshrink tool?

Thanks!

I just scaled it up so it was gigantic. There's a bug in that code that doesn't seem to work right at smaller sizes. Make it big and it works just fine, then scale it all back down :)

-Chilton

jameswillmott
03-01-2013, 09:23 PM
Chilton is right, scaling it helps avoid some precision errors at small scales. It also helps to offset the shapes slightly, mathematically perfect models tend to cause problems.

JonW
03-01-2013, 09:54 PM
It worked here.

After, I redrew the discs so the points had a 2mm spacing along the wire. The heatshrink was a neater result.


Edit: I think it's better to heatshink a 2mm length of wire (1 poly length along the wire) then Array "X" number of times to get a 180 degree twist. It's much neater, easier & there is no cleaning up afterwads.

jeric_synergy
03-02-2013, 01:08 AM
Chilton is right, scaling it helps avoid some precision errors at small scales. It also helps to offset the shapes slightly, mathematically perfect models tend to cause problems.
That's the kinda thing that should be added to the dox.

jwiede
03-02-2013, 01:44 AM
That's the kinda thing that should be added to the dox.
These are visibly large objects, roughly cm-scale, and the scale for many to most 3D genres. This scale shouldn't be more than 100x / 1/100th the "standard internal representation" scale of LW itself, unless the internal scale used is seriously mis-calibrated. If LW's really encountering fp calc precision issues already with cm-scale objects, that sounds like a serious issue given just how much 3D work occurs in that scale. If LW were running into fp calc precision issues in um- or nm-scale, I'd be understanding, but it shouldn't be hitting them at cm-scale.

Lewis
03-02-2013, 02:17 AM
It shoudl be fixed anyway, it's not like this mesh is in nanometers.

gerry_g
03-02-2013, 05:03 AM
interesting discussion, would never put a sleeve on a cable this way would have made a custom profile for it and swept it and twisted it along with the original cables, anyway when I tried it (heat shrink that is) just toggling modes without scaling or any form of geometry clean up made it work for me, when I changed the mode from Normals to Closest Point got the same results as every one else, anyway to hell with the bugs the more I use 11.5 the less inclined I am to revert to any other version, it's starting to feel indispensable

ConjureBunny
03-02-2013, 07:48 AM
Yeah, 11.5 has its share of bugs, but I can say with confidence that some of these new tools are good enough that I've completely changed my approach to critical parts of modeling. I don't think I can go back now.

Also, I realized too late that I can't edit forum posts after some amount of time. Earlier when I said there was a bug in that code, please know that I do not have access to that code, and do not know why it's not working. It might not be a bug. Who knows. Not me. It just didn't work the way I expected it to at smaller sizes.

NT will fix it, I'm sure of that.

-Chilton

Tony3d
03-02-2013, 08:02 AM
I still can't believe no one can reproduce my Juielenne 2 problem. On my new Mac Pro 3.06 12 core under 10.8.2, it does nothing when working in polygon mode. works fine in point to edge modes though. has any other Mac person run into this? could it be a bad download?

ConjureBunny
03-02-2013, 08:06 AM
I still can't believe no one can reproduce my Juielenne 2 problem. On my new Mac Pro 3.06 12 core under 10.8.2, it does nothing when working in polygon mode. works fine in point to edge modes though. has any other Mac person run into this? could it be a bad download?

OH FINE. I'll go try that, too :)

Tony3d
03-02-2013, 09:14 AM
Let me know what you get.

JonW
03-02-2013, 09:52 AM
I've gone mad with a 23 GB file on my box, since I've been running 11.5, I don't want to go back. Not that I have covered every scenario of scenes by any stretch of the imagination. But as far as crashing I'm impressed so far.

The Axis translate for my purpose is seriously useful.

Sorry to get off topic but Lightwave have surpassed itself, & together with LWCad it's it's a killer!


Back to Heatshrink:

would have made a custom profile for it and swept it and twisted it along with the original cables

Have to agree here, Heatshrink in not the right approach for this object, far quicker, simpler & basically no cleaning up to quickly knock up the object. All this cable making is urging me to update my Van Den Hul cables for my KI Pearl amp!

KevinL
03-02-2013, 11:06 AM
Janitor?

Tony3d
03-02-2013, 11:29 AM
If you apply it before the twist it works great.