Page 27 of 28 FirstFirst ... 1725262728 LastLast
Results 391 to 405 of 416

Thread: 1001 ways to use LW and Blender together

  1. #391
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    17,397
    Tried some particle collision effectors in lightwave to spawn particles, particle scan to mdd, and it works nicely.
    Still have some research to do wether or not it is not worth it, or if I will be able to utilize the particle effects in lightwave to such extent that it fills in what is harder for me to do with particles alone in blender.
    You can spawn particles with the older legacy collision system on any other object it hits, or use particles themself to die out and create new particles that could serve as explosion creation, or dust kicking up.

    Hereīs some settings to keep in mind when you apply your blender particles on top of the mdd particles from lightwave, the blender particles should serve as replacement and only follow the mdd particles so fluids can be generated inside of blender.
    Not ready to show the results yet I want better scene case setups before I do that

    Pipeline lightwave particles to blender...(mdd particles are not recognizable as particles in blender, you just use them as placeholders and replace them by overlaying blender particles on to them)

    - Use Dp particle scan, and save as Mdd.
    - Check save lwo
    - Import the lwo saved file from the particle scan dialog.
    - Use mesh cache modifier, not importer.
    (importer brings on shape keys)

    - Set cache to forward up axis to +z and -x and flip axis to y.

    Particle settings..

    - Emission number, try match that of lightwave.
    - Start:0
    - End:0
    - Emit from verts.
    - Physics:no
    - Render billboard scale 0,378 (just for visual opengl feedback to see if blender particles lines up with the mdd particles=

    Apply quick smoke, change flow from smoke to fire and smoke, change source from mesh to particle, and select the blender particle system..save scene and play timeline.
    You most likely need to turn on high resolution and smoke adaptive domain.

    Erikals covers the basics for the export and import of mdd, but this dummy above is for the appliance of blender particles on top of that.
    https://www.youtube.com/watch?v=suZWM0oscqU

  2. #392

    Cool, nice to know that it works.

    LW vidz   DPont donate   LW Facebook   IKBooster   My vidz

  3. #393

    Quote Originally Posted by Marander View Post
    My opinion / experience with various apps and plugins:

    RealFlow - I don't have it. Hmm currently 50% off... But I don't need it and I remember some kind of node locked and online licensing.
    I just read over at the Realflow foum that all the engineers left, so people are jumping to Houdini etc instead.
    http://realflowforum.com/viewtopic.php?f=72&t=12013

    so...
    LW vidz   DPont donate   LW Facebook   IKBooster   My vidz

  4. #394
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    17,397
    Quote Originally Posted by erikals View Post
    Cool, nice to know that it works.

    Yes, you could use it when parenting or linking collision objects..so when these hits break these pieces, at the very moment they hit..it will spawn particles in that area, depending on the size of the collision object, you would only need an initial burst, and that will serve as particle generated fluid emission..and finalized as explosive fluids in blender.

    And itīs also possible to use flocking for some cool motion and the same there apply collision objects on top of that, I think.

    Old pfx, not really tested the full workflow, but it should work for generating particle burst on hits if done as I mentioned above, and there shouldnīt be a problem with that serving as mdd and particle replacement and fluids in blender.

    will try some of that upcoming week I think.
    Got some liquid stuff to deal with as well.
    got some quality rendering issues I need to deal with in lightwave..and a lot lot more.

    I think I need to prep for the winter with a 3D schedule as we talked about experimenting without goals or visions or a plan.



    Even though I experimented with gas solver and some very interesting things in there, itīs just to crude in terms of UX, framework..speed and quality and setting up the shaders etc, far easier to work with the fluids and fire and smoke in blender.
    TurbulenceFD I always had crash issues with, extremely slow curve shading handling including crashing, extremely slow multiple scattering calculation, and the fact itīs is itīs own seperate shader and not pbr.
    It would also mean bigger cost for TFD, and you would probably need to get Octane as well for a decent experience, or a super cpu equiped computerf.

    That I would only do if I worked proffesionally, had the money or got sponsored to invest in it, so currenly fluids fire and smoke in Lighwtave is out of the question, unless something really interestin ghappens with the stagnated TFD version, but that isnīt enough either, it needs to be clear where lightwave will be as well.

    Apart from that, TFD can be nice to work with, calculation speed beats blender fluids, some noise controls and the fire shader is easy to work on as well, though the lack of PBR, it can most likely deal with larger cache files as well than blender does.

  5. #395
    Registered User
    Join Date
    Jun 2014
    Location
    Right here
    Posts
    1,918
    Quote Originally Posted by erikals View Post


    I just read over at the Realflow foum that all the engineers left, so people are jumping to Houdini etc instead.
    http://realflowforum.com/viewtopic.php?f=72&t=12013
    Ouch...! Didn't know they have this issue. Good find, thanks for the info.

  6. #396
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    7,221
    Quote Originally Posted by erikals View Post


    I just read over at the Realflow foum that all the engineers left, so people are jumping to Houdini etc instead.
    http://realflowforum.com/viewtopic.php?f=72&t=12013

    so...
    To be clear, that's a (speculative) statement from a customer, not an official statement from NextLimit, but there's no question standalone RF development has slowed.
    Last edited by jwiede; 10-26-2020 at 02:37 PM.
    John W.
    LW2015.3UB/2019.1.5 on MacPro(12C/24T/10.13.6),64GB RAM, NV 980ti

  7. #397
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    7,221
    Quote Originally Posted by prometheus View Post
    That I would only do if I worked proffesionally, had the money or got sponsored to invest in it, so currenly fluids fire and smoke in Lighwtave is out of the question, unless something really interestin ghappens with the stagnated TFD version, but that isnīt enough either, it needs to be clear where lightwave will be as well.
    Realistically, I think Jascha's just hitting the practical limits of what the (legacy) volumetrics LWSDK can support. I doubt that can improve unless/until NT/V provide LWSDK support for "new volumetrics" (f.e. currently-absent lwvolume.h from C/C++ LWSDK), if even then. Given the uncertainty of LW's future, even if NT/V did somehow* release "new volumetrics" LWSDK support, it's far from clear there's any significant RoI to be had from further investment in LW TFD development (esp. compared to C4D TFD development). No harm to keeping hope, but odds of improvement seem long, at best.

    *: It appears Tindall was the dev for OpenVDB and related volumetrics work, so... (shrug)
    Last edited by jwiede; 10-26-2020 at 03:03 PM.
    John W.
    LW2015.3UB/2019.1.5 on MacPro(12C/24T/10.13.6),64GB RAM, NV 980ti

  8. #398
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    17,397
    Quote Originally Posted by jwiede View Post
    Realistically, I think Jascha's just hitting the practical limits of what the (legacy) volumetrics LWSDK can support. I doubt that can improve unless/until NT/V provide LWSDK support for "new volumetrics" (f.e. currently-absent lwvolume.h from C/C++ LWSDK), if even then. Given the uncertainty of LW's future, even if NT/V did somehow* release "new volumetrics" LWSDK support, it's far from clear there's any significant RoI to be had from further investment in LW TFD development (esp. compared to C4D TFD development). No harm to keeping hope, but odds of improvement seem long, at best.

    *: It appears Tindall was the dev for OpenVDB and related volumetrics work, so... (shrug)
    The proper source to ask that question and get answers from is Jascha himself, I tried quite some time ago..but didnīt get any response on my questions, had a couple of them..but nada, so I couldnīt trust in that development.
    And seeing not much updated for lw isnīt making the trust better, along with those who left that did the vdb gas solver etc.

    So yes, I donīt think thereīs gonna be any good news for any fluid tools in Ligthwave unfortunately, thus I try a lot now to cover what I want and need in other software.
    Embergen will be on testing the maybe within two weeks I think, along with Houdini..and storm may also be something to look in to, and my never endin trials with blender fluids.

    Mantaflow in current state sucks in my opinion, ivé tried to get answers to issues with it over at blenderartist.org...but no one seems capable of answering, either they are not using it much..simply do not now the answer or something, I just find it odd that I donīt get answers within a few days, maybe I was spoiled over here in these forums.

  9. #399
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    17,397
    Fitting for the thread?

    paint3d brain model modified, and reduced (too much though)
    AW bridge tools to merge the connections in just a few seconds, send to blender for faster render and better subsurface material.




    Click image for larger version. 

Name:	Making the Connection_Bridging Lightwave & Blender.jpg 
Views:	207 
Size:	68.9 KB 
ID:	148953

  10. #400
    Registered User
    Join Date
    Jun 2014
    Location
    Right here
    Posts
    1,918
    Quote Originally Posted by prometheus View Post
    Fitting for the thread?
    No, that's how I do it sometimes too with various apps, use the best of each.

    Yesterday I modeled (or at least started) a SciFi bridge in Blender using BoxCutter, KitOps and HOps and then took it over to Cinema for applying Symmetry and Cloner Generators.

    Worked great.

    Maybe I'll export it and render in LW with Volumetric Fog.

    Great to use the tools of different apps where it makes sense.

  11. #401
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    17,397
    Quote Originally Posted by Marander View Post
    No, that's how I do it sometimes too with various apps, use the best of each.

    Yesterday I modeled (or at least started) a SciFi bridge in Blender using BoxCutter, KitOps and HOps and then took it over to Cinema for applying Symmetry and Cloner Generators.

    Worked great.

    Maybe I'll export it and render in LW with Volumetric Fog.

    Great to use the tools of different apps where it makes sense.
    Great

  12. #402
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    7,221
    Quote Originally Posted by prometheus View Post
    Fitting for the thread?

    paint3d brain model modified, and reduced (too much though)
    It was okay for some stuff, but it kind of breaks down here, because it's obvious the darker red shaded areas aren't the "material folds" they're supposed to be -- and the bridging somehow made it even more obvious they're supposed to represent folds, but don't. Bridging ends up looking downright odd, kinda tesseract-y w.r.t. to the "not-folds" (difficult to explain).

    Just mentioning, because it's actually an interesting visual result: The optical illusion of the shaded folds kind of "fights" with the bridging and your eyes attempting to perceive 3D/perspective. I'm always intrigued by these cases where our brains produce multiple unsuccessful solutions for visual relationships, then let the "conscious agent you" examine all of them, and choose which to accept/perceive. The level of cognitive processing and abstractions needed for something like that is just scarily complex, but then so is whatever "conscious agent you" represents, so... (shrug)
    Last edited by jwiede; 10-29-2020 at 04:48 PM.
    John W.
    LW2015.3UB/2019.1.5 on MacPro(12C/24T/10.13.6),64GB RAM, NV 980ti

  13. #403
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    17,397
    Quote Originally Posted by jwiede View Post
    It was okay for some stuff, but it kind of breaks down here, because it's obvious the darker red shaded areas aren't the "material folds" they're supposed to be -- and the bridging somehow made it even more obvious they're supposed to represent folds, but don't. Bridging ends up looking downright odd, kinda tesseract-y w.r.t. to the "not-folds" (difficult to explain).

    Just mentioning, because it's actually an interesting visual result: The optical illusion of the shaded folds kind of "fights" with the bridging and your eyes attempting to perceive 3D/perspective. I'm always intrigued by these cases where our brains produce multiple unsuccessful solutions for visual relationships, then let the "conscious agent you" examine all of them, and choose which to accept/perceive. The level of cognitive processing and abstractions needed for something like that is just scarily complex, but then so is whatever "conscious agent you" represents, so... (shrug)
    I donīt know who did the brain or if it was from sketchfab or lmported from paint3D, have to look back on that.
    The deforms of the brain and the cavities isnīt lining up exactly with the texture and itīs folds.

    Yes, I understand the visual is a bit odd.

    the original mesh was around 4858 polys and I remeshed reduced to 966 polys with lightwave reduce plus, had I opted for a Real showcase I would have kept a higher res of course, would have been more work to merge smaller polygons before the fuse/bridging that way though, and the purpose was just to show the bridging a bit.
    Didnīt do any texture/Uv adjustment or painted in anything new for the bridging.
    I could have been more careful of which polyface selection I had choosen on one brain in relation to what polyface selection I had on the other brain before hitting the bridge, didnīt do any tweak on the handles that much either.

    The AW bezier bridge is cool in the way it curves and bends the bridge in a slight manner depending on the direction from the polynormals, so you can in one button go get a twisted turn by just applying it.
    You should use both spline mode on and off depending on the distance and faces you select, if you have tight distance between the polyfaces you want to bridge, I would recommend using spline mode, if you have a long distance and
    faces that should make large twist, turn it off.

    If you were to make a cup though or a handle from a slight curved faces, and you try to bridge it straight out, You need to select those face polys and make sure those normals are aligned straight to the direction you want the bridge starting out from or you would get a sheared start on it.
    The newer bridge tool in lightwave that is native, you donīt have to worry about that, and you can also select multiple faces which you cant with AW bezier bridge..you have to merge the multiple faces first, so it is lacking there a bit, but
    on the other hand itīs much cooler with how it bridges the faces and the control handles you have.

    The subsurface in blender was random walk, which yields a more transparent look on the thin bridging parts, using Christensen Burley would maintain the texture a bit more.


    straight viewport rendering in blender cycles without noise filtering, noise filtering in Krita instead.

    using the two software with the funcgtions and tools where it does it best, In lightwave that is the bridging action I wanted here, and for subsurface quality..I had to go with blender and also with renderspeed in mind, I also set up a little animation orbit with camera..and here is where I would work with Lightwave instead and camera targeting rather than track to and fiddling with blenders axisy system to get it right, but that would have ment exporting it all..I could have done that but anyway faster just to send the object and align the camera up in blender ..maybe.

    The workflow of bridging is quite fast, I just select one polyface on one mesh, and one on the other, then my fingers are always over the ctrl shift buttons and opens the panel of my favourite used tools and select bridge, prefer that rather than having to change to a shortcut press, something about that which makes it more fluent to me when bridging.
    then do the same on a couple of others and it goes quite fast to produce connections like this.

    I should get me 3rd powers metamesh though to expand on this with other premade structures and fusing together, mixing with bridging segments when I need to.








    Christensen Burley subsurface mode, not the same translucent look on the thin parts as the main image I posted.

    Edit..Oh, I had one little sphere with emission material, lighting up the middle part a bit, could have used ordinary lights also, but I was experimenting with some other stuff initially.



    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	brain subsurface Christensen Burley.jpg 
Views:	55 
Size:	176.7 KB 
ID:	148958   Click image for larger version. 

Name:	brain reduced mesh & bridging.jpg 
Views:	56 
Size:	293.5 KB 
ID:	148957   Click image for larger version. 

Name:	brain original mesh.jpg 
Views:	54 
Size:	442.5 KB 
ID:	148956  

  14. #404
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    17,397
    And again with hair, you can also just add fibers to your object directly in layout, make some styling, then polygonize it, sides 1 and 100% flatness, then a mesh will be created to save as lwo
    load it in to modeler and select edges, copy edges, switch back to polymode and delete those polys, then paste the edges back again.
    Then save out to obj format.

    You need the mesh you used to create the fiberfx from and then the strand result, just save them out as obj, import to blender 2.79.3 and use the hairnet plugin, the hairnet can be found in
    the particle tab.
    Make sure you select the strands first then the mesh and then use hairnet from fibers.
    with child particle display on, you have to turn down the radius to a very small value, and also possibly change roundness.

    There is a newer version of hairnet mainly for 2.80, I tried that in 2.83 though since I do not have 2.80 installed, didnīt work the first times..but I got it working oddly with some fiddlings, but I think itīs not optimal..have to check that more.

    Also, you can of course also comb this with particle edit in blender, since you should remove the mesh you used for the hairnet conversion, and just jump in to the particle edit for that generated hair...so no difference from standard blender particle hair ..since it is just that once converted.

    Sorry...not time right now for a better hair model than the standard sphere
    And as I have mentioned, the hair shader in later 2.83 is better than what I used in this image.

    Also to note, cycles 4.39 seconds to refine in this case, lightwave with some 1000 fibers..over 84 seconds to refine, and then it isnīt still half as good as the blender quality, amount of fibers in blender is a bit unclear, 200 main strands with 50 in value of the display for the strands...have to check that.








    Click image for larger version. 

Name:	fiberfx.jpg 
Views:	37 
Size:	274.0 KB 
ID:	148976

    Click image for larger version. 

Name:	converted with hairnet_no childstrands.jpg 
Views:	37 
Size:	259.9 KB 
ID:	148977

    Click image for larger version. 

Name:	with childstrands.jpg 
Views:	37 
Size:	275.5 KB 
ID:	148978

  15. #405

    Also to note, cycles 4.39 seconds to refine in this case, LightWave with some 1000 fibers..over 84 seconds to refine, and then it isn't still half as good as the blender quality, amount of fibers in blender is a bit unclear, 200 main strands with 50 in value of the display for the strands...have to check that.
    yeah, rendering hair in LightWave takes quite long. 
    LW vidz   DPont donate   LW Facebook   IKBooster   My vidz

Page 27 of 28 FirstFirst ... 1725262728 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •