Page 3 of 21 FirstFirst 1234513 ... LastLast
Results 31 to 45 of 304

Thread: Seriously NewTek... This is your best moment for LW2020

  1. #31
    Quote Originally Posted by Imageshoppe View Post
    I thought a little more about my reluctance to move on from 2015 even though I purchased 2018 and 2019 promptly on release, and will do the same with 2020 or 2021 or Viz Whiz or whatever it will be called...

    ... The change up from 2015 to 2018 required all the effort of adapting your workflow to a revolutionary new product, but without the actual revolution happening. What I mean is that there are a few, new, nice features that seemed worth the upgrade, but haven't yet been actually worth the whole-scale change over to a new object format, the changes in surfacing, and the overall slowdown of the render itself.

    It doesn't mean I don't fire up 2019 and render now and then, but ONLY for things that 2015 can't do. As example, the volumetric clouds from this shot, which took every trick in the book over three computers to get rendered in a timely fashion in 2019, but the actual city structure had to be 2015 rendering due to the excessive time needed to get a clean, smooth image in 2019 with little aliasing and noise. It may not be as realistic as the PBR workflow of 2019, but I'll trade "nice" over "ratty" any day (there is some aliasing due to the downsample to the 1080p mp4 from Vegas Video) but the original is nice and clean...

    http://files.datausa.com/imageshoppe...ORBIT_HIGH.mp4
    Part of the reason I had to move on from 2015 was that 2015 would run really buggy on my new hardware unless I booted up in 16 Core mode. I always want to try and take advantage of the latest features too so that was the other part of it. A bit of a painful transition but once your on the other side it's clear sailing.
    Threadripper 2990WX, X399 MSI MEG Creation, 64GB 2400Mhz RAM, GTX 1070 Ti 8GB

    https://www.dynamicrenderings.com/

  2. #32
    Quote Originally Posted by Nicolas Jordan View Post
    Part of the reason I had to move on from 2015 was that 2015 would run really buggy on my new hardware unless I booted up in 16 Core mode. I always want to try and take advantage of the latest features too so that was the other part of it. A bit of a painful transition but once your on the other side it's clear sailing.
    Yeah, that's a BIG reason to abandon 2015 full stop! I think I remember you saying something about that when you first got your 32 core. I settled for the 1950X at the time, which was a first gen Threadripper and a bit of a value/deal. Interestingly, I recently looked into upgrading to your 2990WX and it still commands a good price, so smart move on your part, as it hasn't depreciated much (well, compared to the consumer chips) over the year.

    Quote Originally Posted by prometheus View Post
    There lies a problem with that approach, and for critics in your clip, the flying building structure doesnīt seem to belong there in the scene in the clouds.
    The surface seem to be reflective, but doesnīt seem to reflect the environment, and some other lighting and shading issues I think.
    Oh, it's perfectly reflecting the environment and the lights are in the right places, here's a quick render where I drop a mirror ball into the scene... I don't think you like my surfacing choices on how/where it's reflective and what I did with color iridescence at extreme angles... that's cool, but the client loves it. There is a BIG problem with the shot becase at the last minute they requested that lower ballast boom weight be added, and this was after the rest of the shot had been locked in stone... unfortunately, the boom weight is actually lower than the top of the clouds! I couldn't re-render the clouds lower in quick enough time, nor roto the boom into the clouds (the point was to see the boom weight in the first place). It creates an optical illusion that once you see it you can never un-see it. So it is as it is... but that's the nature of deadline based production work sometimes. There's 17 other minutes to finish so you have to move on...

    http://files.datausa.com/imageshoppe...irror_ball.jpg

    One final note on that image... again, to get 2019 to that level of smoothness with the fine lines and lack of noise/grain in the reflections is quite a burden. This frame in 2015 rendered in 89 seconds thanks to pre-rendering the lower volumetric clouds and adding them into the scene by camera front projection... and there's 1799 frames more to render for the shot, so everything is always a balancing act.

    ...also revisit the link with the clip, as I went ahead and did a better encode at higher rez and it goes on for the full 1 min run-time.

    Bringing this back to LW 2020... there is so much that people want/expect from LW right now... I don't know how it can possibly satisfy. But for me, it has to be a refocus on actual animation aspects, if there's noise reduction it has to be temporal as well and look good across frames... or maybe CPU/GPU hybrid rendering... something like that would make the rest of the transition effort from 2015 be worthwhile.

  3. #33
    Registered User
    Join Date
    May 2019
    Location
    cologne
    Posts
    86
    ...because Modo 14 is out.......and now....itīs your turn.

  4. #34
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    16,751
    Quote Originally Posted by kievi View Post
    ...because Modo 14 is out.......and now....itīs your turn.
    The marketing trick of recent seem to be .."Itīs a game changer" in many releases out there today, discarded is state of the art, and out of the box.
    A short look at the 14 version, some things they already got in blender...what I could see useful is procedural vertex maps and the mesh fusion improvements perhaps.

    But game changer?? really.

    And when are they going to get a fluid system?
    There was talk about Jascha porting it to Modo in 2014, and he said it was planned, but itīs been 6 years and nothing?

  5. #35
    Video Junkie Vong's Avatar
    Join Date
    Feb 2003
    Location
    Eagle Mountain, Utah
    Posts
    320
    Keep in mind, when speaking of MODO 14 being a game changer, you'll need to wait until the end of the year when all the 14 features are implemented!
    Curtis Rhoads

    Desktop: Win10 x64 | i7 3.4GHz | 32GB RAM | GeForce GTX970
    Laptop: Win10 x64 | i7 2.3GHz | 8GB RAM | GeForce GT650M
    Desktop: iMac | i7 3.5GHz | 8GB RAM | GeForce GTX 780M

  6. #36
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    16,751
    Quote Originally Posted by Vong View Post
    Keep in mind, when speaking of MODO 14 being a game changer, you'll need to wait until the end of the year when all the 14 features are implemented!
    Oh...premature feature release, always disappointing all this build up, then it comes too early.

  7. #37

    Modo... meh...



    + many other things.
    LW vidz   DPont donate   LightWiki   RHiggit   IKBooster   My vidz

  8. #38
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    16,751
    Quote Originally Posted by erikals View Post
    Modo... meh...



    + many other things.
    They have a strange policy of making it not so easy to evaluate the software, several times Ivé complained about their demo versions and policy, wish it were like lightwaveīs with a decent discovery version..as soon as the 30 day trial is over, it seems you have to wait for the next Big release update.
    I Just hate it.

    Lightwave & Houdini rocks in that area.

  9. #39

    if adding something, it is either B or Houdini for me now.

    with Houdini i tend to go all "nodal" so i'm less of a fan, the nodal substructure is just too much.

    with DeepRising fluids seem to come along inside LightWave. NewTek should support him.

    a shame about Houdini, i spent tons of time testing it. but in the end it was just too complex for a One Man Band...


    LW vidz   DPont donate   LightWiki   RHiggit   IKBooster   My vidz

  10. #40
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,879
    Quote Originally Posted by erikals View Post
    Modo... meh...



    + many other things.

    Last edited by jwiede; 03-19-2020 at 02:05 PM.
    John W.
    LW2015.3UB/2019.1.5 on MacPro(12C/24T/10.13.6),64GB RAM, NV 980ti

  11. #41

    LW vidz   DPont donate   LightWiki   RHiggit   IKBooster   My vidz

  12. #42
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    16,751
    Quote Originally Posted by erikals View Post
    if adding something, it is either B or Houdini for me now.

    with Houdini i tend to go all "nodal" so i'm less of a fan, the nodal substructure is just too much.

    with DeepRising fluids seem to come along inside LightWave. NewTek should support him.

    a shame about Houdini, i spent tons of time testing it. but in the end it was just too complex for a One Man Band...


    Giving up on Houdini for the moment then?
    Blender isnīt as deeply nested in terms of node requirement workflow, Lightwave seem to be going on the nodal structure much more which I donīt like...while houdini tries to wrestle out of the nodes ..at least to provide tools properly wrapped around it with as much as direct acess to the settings...but still has a long way to go.

    For fluid smoke ..blenderstill have a bit left to work on I think..especially for dealing with highres simulations, I wasnīt happy about how mantaflow works..compared to the older fluid system.
    and for fluid smoke in Lightwave natively with gas solver..still a long way to go.
    And for fluid smoke plugins...TFD, well ..Jascha takes very long time with his stuff, I was actually amazed to see him come up so fast with what he did initially for Lightwave, then the development speed just went up in smoke litterally.

    Houdini or cinemaversions of tfd perhaps, but that is not my priority either..
    Perhaps switching to 3d max, get more expensive tools and include thinking particles and tyflow...just to follow Alan McKays stuff...Nah, too expensive, too much time to re-learn..and re-learn a software UI I never liked anyway.

    Kind of in a limboland for years where I do not get any satisfaction with any of the fluid systems...I may not be able too, and have to narrow it down someday and accept whatever shortcomings it may have.
    Realisticly ..there will always be issues.

  13. #43

    Giving up on Houdini for the moment then?
    unfortunately...    

    Perhaps switching to 3d max,...
    Never.

    Kind of in a limboland for years
    yeah, selecting the right choice(s) is trickier than one would think...    
    LW vidz   DPont donate   LightWiki   RHiggit   IKBooster   My vidz

  14. #44
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    16,751
    Quote Originally Posted by erikals View Post

    unfortunately...    


    Never.


    yeah, selecting the right choice(s) is trickier than one would think...    
    Iīll get there someday.
    Clip...to low resolution for the delivered vid, and too low grid resolution for the simulation, no additional noise method, explosion also going too fast etc.
    So quite a bit higher in resolution and perhaps some slight noise method added, lower the initial force/speed of the explosion, and add a tank vehicle, some background environment elements.


  15. #45

    looks nice, and yep, needs an upres.  
    LW vidz   DPont donate   LightWiki   RHiggit   IKBooster   My vidz

Page 3 of 21 FirstFirst 1234513 ... 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
  •