Jump to content


  • Content count

  • Joined

  • Last visited

  • Days Won


Everything posted by malexander

  1. Background image

    It's greyed out because the camera is currently defining the background image. If you lock the view to the camera then it'll enable these options, and any changes you make will be applied to the camera Background Image parameter.
  2. Rendering Normal Maps?

    You probably need to flip the green channel (1.0 - G). There's 2 styles of normal maps, GL and DirectX, due to the fact that window Y=0 is at the bottom in GL, and the top in DirectX (matching X11 and Win conventions).
  3. Is AMD potentially risky? (Threadripper)

    I think it's worth noting that any new CPU architecture is inherently risky. AMD's offering has NUMA issues and Intel's heat issues (due to poor packaging of the heat spreader). If you have any issues, I'd suggest dialing back any overclocks. Rendering a sim or sequence isn't any faster if it crashes midway through.
  4. Compositing- increase canvas area

    Yep, Window COP. Also, the Border COP will give you some additonal options about how you want that area filled, if you need that.
  5. Controlling Focus Distance

    You can also use an Object CHOP to get the distance between two objects, then put a chop() expression in the focus parameter to fetch the result.
  6. How to set alpha channel with function

    Channel Copy with the following settings will copy the luminance to A and set C to black: Target: New Plane: A Source: Input 1: C Target: C Source: Set to 0 (black)
  7. Is AMD potentially risky? (Threadripper)

    Higher RAM speeds do increase performance slightly. It's worth getting if the markup on the price is <10%, but that's rarely the case. Also it's harder to get fast RAM in large DIMM packs that you'd need to populate all the slots (4-8). If you have to choose, always go for more RAM over faster RAM. You can see the effect of RAM speed, and NUMA, in the first page of this article: http://www.anandtech.com/show/11726/retesting-amd-ryzen-threadrippers-game-mode-halving-cores-for-more-performance
  8. Take the 1080Ti. You won't regret it. The iMac, on the other hand.... (^^^)
  9. Well for one, it's really hard to display ptex in realtime using any of the current APIs. The individual textures need to be assigned to an atlas as DX and GL don't support infinite numbers of texture samplers (in most cases), and that texture atlas setup isn't exactly speedy especially if you're still modelling.
  10. Is AMD potentially risky? (Threadripper)

    The Zen cores are arranged in modules of 4. Intel cores are generally paired together with shared L2, though I think the latest iteration of Skylake-E removes that (7xx0 series). It's pretty common practice to have cores share some resources to keep power requirements down. Ryzen has two of those modules on a die with a memory controller. Threadripper has 4 modules with 2 memory controllers, and that's where the NUMA (non-uniform memory arrangement) and proper OS scheduling comes into play. The first 2 modules have access to one bank of memory, and the other 2 modules have access to the other bank. If a core from one module needs memory from the other module's bank, there's an extra hop to access the memory. That's the "non-uniform" part, since RAM latency can vary based on its physical location. Accessing RAM is already pretty slow, which is why CPUs have large L3 caches, and use SMT (aka Hyperthreading(tm)) to hide the RAM access latency. Thread stalled on a memory request? Switch to the other one that's parked on the core and continue crunching numbers. The OS scheduler is also responsible for keeping threads on one module or the other if possible, so these days NUMA doesn't have quite the hit that it used to on the older multi-socket servers. That's why sometimes a software or firmware update is needed for new CPUs.
  11. Houdini 17 Wishlist

    Before VEX Houdini had TOPs, which were Texture OPs. But the new TOPs is not a reincarnation of them. Try again
  12. OpenCL simulations on ADM GPU

    Houdini either uses OpenCL 1.1 or 1.2. OpenCL 2.0 doesn't expose any features which Houdini needs right now. Single precision is also more important at the moment, and as Atom says, only 1 compute device is currently supported.
  13. render EXR colorspace

    EXR is a linear-space format. No conversions are done when writing from mantra (which working in linear natively) to EXR.
  14. node color conventions

    COPs has a standard color scheme, which was implemented with the original design: Generators are green VEX operators are purple Operators which don't affect image data are beige (timing operations, control flow, etc) Collapsible Color Corrections (precursor of Compile SOP networks) are blue The last three had special meanings with regards to cooking, which is why they were colored by default. VEX operators could bind planes named the same as VEX function parameters, Timing operators didn't take long to cook because they didn't modify image data, and Color Correction nodes would be collapsed into a single operation if there were no other non-color correction operations between them, saving memory and improving performance. But other than that, I don't believe there are any standard color schemes in Houdini.
  15. Crowd Agent not showing in viewport

    Crowd display unfortunately doesn't work on Mac because the GLSL shader just... doesn't draw anything (?!). The same shader works on other platforms. If you switch to wireframe you can at least see the rigs.
  16. AMD GPU pro drivers stability

    It depends on the addressing mode of OpenCL - Nvidia took a really long time to add 64b. If "Device Address Bits" is 32, it can only access 2GB. It needs to be 64 in order to access the full memory on the card. This was fixed a couple of years ago in the Nvidia proprietary drivers, though I'm not sure what their state in MacOS is .
  17. It's probably good to get into the habit of thinking that anything you do at work, or with work assets, is the property of your workplace. If you keep work and personal materials separate, you'll never need to worry about losing your personal stuff. And more importantly, you won't have to worry about being accused of any sort of theft, leaks, or using work assets for personal gain. Best to play it safe and keep them well separated.
  18. AMD GPU pro drivers stability

    Mac is in its own little world, you get whatever driver that particular OS version gives you. While some vendors like Nvidia allow you to download the back-end driver part, the OpenGL interface remains the same regardless of vendor or driver. So yes, there's no consumer or pro driver variants, and Apple in general seems to use consumer grade parts (though sometimes semi-custom). I guess there's no reason to offer pro parts when the GL interface doesn't allow access to pro features like quad-buffer stereo, genlock, etc. Not sure if 10b framebuffer support is the exception, as you'd think Apple would want to push their displays if they support 10b RGB color. I had a hard time even finding the bit depths of the Apple monitors listed, which might suggest that they are currently 8b.
  19. AMD GPU pro drivers stability

    I'm using AMD FirePro version 16.50.2001 and it's quite stable (w/ W8000; hgpuinfo reports 21.19.234 so who knows what the actual version is). There is a bug in a later version of the drivers which AMD has on their radar which breaks vertex selection in the UV viewport (May and June drivers of this year). Not sure if this would also be in the driver for the Vega-based Radeon Boggle-edition, but there is a good chance it may be, as that would require a very current driver. I think that's the only known issue though.