Cheetah3D User Forum  


Reply
 
Thread Tools Search this Thread Display Modes
Old
  (#1)
Helmut (Offline)
Erfahrener Benutzer
 
Helmut's Avatar
 
Posts: 1,356
Join Date: 18.01.2010
Location: Vienna
Default Particle Mesh Problem - 03.07.2018, 11:40

* I am experiencing a weird problem here with a particle mesh:
* The model shows specific clusters of mesodermal cells, somites, which speciate during embryogenesis and in later stages form vertebrae, ribs and muscle-tissue. In simple terms these are cellular blobs paired along the axis of the proto-spine.
* An obvious method for modelling this is a particle mesh on a curved invisible backbone. For unknown reasons, the two central blobs do not align properly. Equally odd: When the symmetry modifier is activated, these central somites are mirrored on a different plane as are the rest of the blobs.
* Does anybody have an idea what is happening here?

* I have also included options B and C, particle spline and chain. These work as expected.
Attached Images
 
Attached Files
File Type: zip 6 Somites 2.jas.zip (29.5 KB, 10 views)
   
Reply With Quote
Old
  (#2)
frank beckmann (Offline)
Erfahrener Benutzer
 
frank beckmann's Avatar
 
Posts: 15,530
Join Date: 12.07.2005
Default 03.07.2018, 12:21

Any reason not to use the "Chain creator" instead?

Cheers
Frank
Attached Files
File Type: zip 6 Somites 4U.jas.zip (34.4 KB, 7 views)


Life is short. Eat dessert first.

Last edited by frank beckmann : 03.07.2018 at 12:32.
   
Reply With Quote
Old
  (#3)
misoversaturated (Offline)
Erfahrener Benutzer
 
Posts: 497
Join Date: 05.05.2016
Default 03.07.2018, 12:40

Hi Helmut!

The "align" function of the particle mesh does only consider those two dimensions of a polygon in which it is extended, the third dimension in which it is flat doesn't allow for proper orientation.

This sometimes results in weird behavior of particles aligned to horizontal polygons as in your example.

I have not properly understood all consequences of this but in my experience there is less trouble with vertical polygons as a guide for particles.

So when I turn your invisible backbone 90 degrees so that it is "standing" instead of "laying" and then adjust the rotation of the red capsules they suddenly align properly.

When you put the guide mesh and the particle system in the same folder you can rotate the folder back to the intended orientation with the particles retaining their proper alignment.

EDIT: For whatever reason the radius of the somite in your file was at 0.0 so that they don't show in 3D view.
I assigned a value for the radius and saved my modified version of your file but when I reopen it the radius again is at zero
Attached Files
File Type: zip 6 Somites 3.jas.zip (15.9 KB, 10 views)


MacBookPro Retina 2.3 GHz i7 16 GB Ram nVidia 750M / OS 10.9.5 / Cheetah3D 7.3 beta / Blender 2.79 / LuxRender 1.6 / PS CS4

Last edited by misoversaturated : 03.07.2018 at 12:49.
   
Reply With Quote
Old
  (#4)
Martin (Offline)
Erfahrener Benutzer
Site Admin
 
Martin's Avatar
 
Posts: 9,166
Join Date: 13.03.2005
Default 03.07.2018, 13:45

I've fixed that problem. It will work properly in the next Beta.

Thanks for reporting.
   
Reply With Quote
Old
  (#5)
Helmut (Offline)
Erfahrener Benutzer
 
Helmut's Avatar
 
Posts: 1,356
Join Date: 18.01.2010
Location: Vienna
Default 03.07.2018, 19:47

* @ the Good Doctor, Martin: Thank you for the pro futuro fix.
* @ Frank B: I do need the precise alignment of them blobs. This is planned as a "single take" video of some 5` duration starting with the the ovum + sperm (= fertilisation) which morphs into a (fairly primitive) proto-embryo. Each take must fit seamlessly into the next stage of embryonal development.
* Screen shots show the morula (= a cluster of undifferentiated cells on day 2) and a simplex embryo (about end of trimester 1). There is no cut in intervening takes. Of course, there are individual takes (about 24), but each take starts with a frame identical to the last frame of the preceding take.
* The currently identical somites - simple place-holder blobs - will transform in a subsequent model into the 7 cervical vertebrae (= neck), the 12 thoracic vertebrae (= rib cage) and so on. This can be achieved, with bit of trickery, using particle meshes.
* Toggling off the alignment check box results in somites which are not aligned to the notochord (= the primitive spinal cord). As a result, further smooth detailing gets complicated.
* @ MisOS: The radius of somites starts at ° = 0 and is incremented in the animation. This is a development of mesodermal cells. Thank you fo for the hint re rotating the backbone. I need to perform some tests to evaluate how this affects later concepts of embryonal development.

* Thank you all for the help in this matter. Muchly appreciated.

* Sorry, this is a bit complicated. Most of us have performed embryogenesis without any problems, but most of us did not have the slightest idea what we were doing
Attached Images
  
   
Reply With Quote
Old
  (#6)
misoversaturated (Offline)
Erfahrener Benutzer
 
Posts: 497
Join Date: 05.05.2016
Default 03.07.2018, 21:21

Quote:
Originally Posted by Helmut View Post
The radius of somites starts at ° = 0 and is incremented in the animation.

OMG I didn't switch the timeline on and checked if it was an animation!

And so I missed the interesting parts of what you have created there!
Checking my file again at least it works also animated

Looking forward to Martin's fix of the alignment problem now ...


MacBookPro Retina 2.3 GHz i7 16 GB Ram nVidia 750M / OS 10.9.5 / Cheetah3D 7.3 beta / Blender 2.79 / LuxRender 1.6 / PS CS4
   
Reply With Quote
Old
  (#7)
frank beckmann (Offline)
Erfahrener Benutzer
 
frank beckmann's Avatar
 
Posts: 15,530
Join Date: 12.07.2005
Default 03.07.2018, 22:37

I still don┤t get why you think the particle system is more accurate than the chain creator. I just took your spline and attached the assets you created and the animation works already without any further work.

Cheers
Frank


Life is short. Eat dessert first.
   
Reply With Quote
Old
  (#8)
Helmut (Offline)
Erfahrener Benutzer
 
Helmut's Avatar
 
Posts: 1,356
Join Date: 18.01.2010
Location: Vienna
Default 04.07.2018, 18:18

* @ Frank: You are probably correct as to particle mesh vs chain creator. It may not make much of a difference in this topology / at this state of embryonal development.
* I simply prefer the particle system because it is more flexible for tweaks and subtle customising.
* In a particle system I can populate the single PM hierarchy with different "types" of cell clusters (rear bit of the skull, cervical, thoracic, etc) and activate sorted mode. A simple copy & paste of subnodes would give me 7 vertebrae for the neck, 12 for the nascent rib cage and so on.
* A PM also has the tag option for fine tuning / animating the standard parameters and variations.
* In a chain I am stuck with a single object and the spline-axis. The Euler angles (heading, pitch, bank = yaw, pitch, roll) are not really useful in this construct.

* The development of an embryo is an amazing patchwork of seemingly chaotic cellular blobs and I need as many degrees in the freedom of the design as C3D and I can possibly handle. Otherwise this project turns into a 9-months pregnancy replete with morning sickness plus intermittent cravings for ice cream with picked onions and dollops of peanut butter
   
Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump



Powered by vBulletin Version 3.6.0
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
vBulletin Skin developed by: vBStyles.com