Welcome to od|forum

Register now to gain access to all of our features. Once registered and logged in, you will be able to contribute to this site by submitting your own content or replying to existing content. You'll be able to customize your profile, receive reputation points as a reward for submitting content, while also communicating with other members via your own private inbox, plus much more! This message will be removed once you have signed in.

goldleaf

Members+
  • Content count

    280
  • Joined

  • Last visited

  • Days Won

    3

Community Reputation

42 Excellent

About goldleaf

  • Rank
    Illusionist

Contact Methods

  • Website URL http://vimeo.com/user866299

Personal Information

  • Name Chris
  1. Yeah more ram/HD space the better, if your motherboard can support it. For your current setup, you can totally get used to using Houdini just fine; obviously keep in mind you won't be winning any benchmarks, but for just getting started for as inexpensive as possible, you're pretty close to having something decent to start with; I would replace your graphics card with a 6GB GTX 1060, for $250 USD (I use one on a 2560x1440 screen, works great). Also, I've found that older 12-16 core (24-36 thread) workstations can be found on eBay for pretty good prices, and are a cost effective way to acquire cpus for heavier sims/renders. Often times these machines can support anywhere from 96GB to 256GB of ram too. I also use Linux Mint. Good luck!
  2. if you need to save disk space, you might find it useful to convert large particle sets / point clouds into Points VDB and saving that VDB primitive to disk in your bgeo. Haven't tested rigorously but seems promising.
  3. http://www.openvdb.org/download/openvdb_particle_storage_2015.pdf
  4. The initial data for DOPs simulations requires .sim files. For FLIP (and maybe some other solvers), technically you could probably use a bgeo as the source, giving it some initial velocity and such based on that frame of a sim; but you're not actually resuming a sim, you're starting a new simulation, and it may not match up very well since it's an incomplete snapshot that doesn't have the whole state of the system like the .sim files would have. *edit* what vtrvtr said
  5. Yeah that's awesome they added it; I switch between left/right handed mice/stylus and backspace its great. There is also a Left Handed Menus option, to move the menus to the left side of the cursor instead of the right side, if you prefer that.
  6. Definitely sounds like a bug, though I haven't hit it; I'd submit it pronto.
  7. Try the Group Expression SOP with: @numvtx > 4 Group features were broken up to be easier to support/expand, and to make it easier to work with groups. Many of the features crammed into the Group SOP were difficult to find/use. Hope that helps!
  8. Yep, that's what I was curious about! Glad to know it's in there! Thanks @Sepu
  9. Thanks @Atomthat's great! Does RedShift allow variations when instancing lights?
  10. I too am curious, its good to see AMD surging in the right direction
  11. That's interesting how much Active Radius made a difference on your system! I wonder what was behind that? Single proc vs dual proc? Or maybe the operating system?
  12. Thanks so much for the conversion @galagast that's great! Here are the specs for the system I ran those tests on (totally forgot to include/mention that before, thanks!): Linux Mint 18.1 2x Xeon X5670 @2.9Ghz 96GB RAM GeForce 1060 GTX I also updated the original post to clarify they were disc lights, not sphere lights.
  13. Here's the scene, and a quick how-to: https://vimeo.com/207214689 http://blog.cerebero.com/post/158088992098/light-instancing-for-mantra https://d.pr/APrK (attached to the post, since droplr is blocked for some folks) instancing_lights_files.tar.gz
  14. For anyone rendering scenes with mantra, and have lots of lights, there have been some amazing optimizations that showed up in the last few days. The scene has 528 disc lights with varying colors: 16.0.504.20 11m 51s (no active radius) 8m 29s (active radius 3) 16.0.535 3m 28s (no active radius) 3m 13s (active radius 3) 16.0.537 1m 18s (no active radius) 1m 9s (active radius 3) I haven't had a chance to try it in 15.5, since the scene was built using several H16 nodes, but if anyone else gets the chance to compare against 15.5, that'd be cool!
  15. Great insights, it's good to hear some experiences from more in-depth testing. Thanks Jason!