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.

Neon Junkyard

  • Content count

  • Joined

  • Last visited

  • Days Won


Community Reputation

15 Good

1 Follower

About Neon Junkyard

  • Rank

Contact Methods

  • Website URL www.neonjunkyard.com

Personal Information

  • Name Drew Parks

Recent Profile Visitors

1,531 profile views
  1. I was also looking at those chips, suspiciously cheap for what you get...I would like to know more about this as well. I'm currently running a (somewhat) outdated dual xeon build, it runs great but slow core clock speed, I'm thinking convert it to a dedicated render/cache slave, build a new faster box and also get into the GPU rendering game. My current mobo (supermicro MBD-X10DRL-I) only has 1 PCIe x16 slot, and I would want to run minimum 2x GTX 1080ti cards, maybe more, but still have a dual xeon system with tons of RAM. Traditionally I think that is a somewhat contradictory ask because it is a server build, and admittedly I really don't know that much about GPU rendering or how to plan a build for it, but I would like to have the best of both worlds- dual xeon 16-24 core machine with 2+ high end cards for GPU rendering as well. Anyone have any advice on that type of build? Windows btw
  2. I can't find the auto-place nodes on creation option in H16...really annoying, is it gone or am I missing something?
  3. Yea I understand that..obviously hotkeys are gone. The major things I am looking to salvage- Galleries Node presets & defaults Orbolt downloaded OTLs/HDAs Shelves
  4. Hey guys, I want to upgrade to H16 but keep all my preferences, shelves, galleries, node defaults, node presets, custom desktops, etc. I have had some problems in the past with hip files getting corrupted likely due to corrupt preferences, bizarre shader errors from old scenes, etc. from just copying the %HOME% folder into the new version. Maya has the same issues, it just makes me nervous Does anyone have the cleanest way to upgrade without copying the whole preferences folder over, like specifically what files you need and what files are likely to create conflict? Same applies to daily builds, seems like a lot of the preferences are stored in the install directory and not the %HOME% directory, I would like to figure out a cleaner way to upgrade without having to redo a lot of saved prefs
  5. that is a great start, thanks!
  6. As the title says I would like to (somewhat) recreate the functionality of nuke;s grade node, in SOPS, as on OTL. I find the color correct pretty unintuitive, and would like the controls nuke offers which I find very easy to use. Obvious controls for black and white point, gain exposure gamma etc. Also the ability to tint any parameter is a big bonus. However I am having a hard time finding info on the math that goes into even basic color corrections, i.e. if you wanted to tint the gamma (midtones) how would you even go about doing that. Haven't found any good rescources on this topic Thanks!
  7. Similar to the H14 workflow of hitting X on the currently selected node in any vop context to create a visualizer node, I want to do the same but connect the currently selected node to the compute lighting Ce output (or a bind export node set ot Ce but that can throw errors) to quickly visualize the color data of the map I am working on without having to do expensive shader calculations. The visualize node is great when working with Cd in SOPS but does not return the same functionality in shaders This is a default hotkey in arnold for C4D, its like alt+w+v or something. I'm sure there is a way to do this with a python script attached to a hotkey. Any ideas?
  8. So this might be a question of mantra settings in general but I am having problems getting rid of indirect reflection noise, it seems like no matter how high my settings this pass is coming out extremely noisy and it is showing up in the final render. I have read through all the docs regarding noise and read through the forums but since mantra changes with every release I don't know what information is still relevant. For instance I read the practical upper limit for pixel samples was 6x6 unless you have heavy DOF or motion blur. I have no idea what extreme settings would be for anything else though like sampling quality on lights or min/max pixel samples. The docs say in some cases you have to increase min ray samples for indirect reflection/refraction noise, but I read elsewhere not to increase that. What is the workflow there, when should this be increased? And how does that affect reflection/refraction quality, which seems to be a global multiplier? Also on the noise removal docs it says as an example 100 max indirect ray samples to achieve a clean image, is that referring to "enable indirect sample limits" rendering parameter on a shader? Is there not a more global way to set that or is that what sampling quality on each light is doing? However increasing that per light doesn't seem to make any difference. I should mention I come from a vray background so that is the workflow I am used to, if anyone can draw any parallels. Also this actual scene doesn't really matter, these are more general questions about newer mantra settings and when to tweak what Quick info about my scene- Still image, no DOF or motion blur, no environment light, just 3 area lights with sampling quality of 9 I do have some fine displacement which might be part of the issue, set shading quality to 3 on that object. Here are my render settings (refl and refr limit at 5, no diffuse bounces, color limit 5). Everything else is default Indirect reflection pass- Here is the comp zoomed in a bit, you can see there is visible noise on the black spheres and the displacement, there is also some square artifacting happening, my guess is the displacement is just too fine for mantra to dice properly resulting in a moire pattern type of effect but that wouldn't explain the square artifacting (that is in the raw exr, is not jpg artifacting) Any help on this would be greatly appreciated
  9. Hey just wanted to point out, not sure if this is a new feature in one of the daily builds but if you load a blank scene in houdini and set it to manual then open your file it will automatically open it in manual mode, no need to fuss with command line
  10. Thanks for the tip about manual update, I will definitely keep that in my back pocket I was able to get the scene stable by watching what the command line was doing when it crashed, and as I was able to open it on another computer with a fresh houdini install I locked a file cache node downstream from the problematic nodes and now it is relatively stable Unfortunately I can't post the file as it contains copyrighted material and whatnot, plus knowing life it probably wouldnt be reproducible on the other end hah. I was just wondering if anyone know for instance, like don't alter the build it HDA's and save them over the original or something. I know so many ways to fix a maya scene like optimize scene size, python node clean scripts, flush bad geo through zbrush etc. To me it seems houdini doesn't have any tricks like this, just the preferences get corrupted and you have to start over, which is annoying
  11. To add to this I opened the scene on another computer with a different local install setup, still crashed, but trashing the preferences worked. However doing that exact same process on my main workstation does not work, still a hard crash even with fresh preferences Also I installed the latest daily build and my graphics card drivers are up to date
  12. I had a somewhat crash-y .hip file, and now houdini hard crashes every time I open it, the error I'm getting on the console is 2724: fatal error: segmentation fault. running windows 7 pro I read through the forums and someone suggested backing up all your prefs and trashing the houdini15.xx folder in the $HOME directory, in users\USER\documents\ which worked; then I copied all my nodes over to a fresh scene (thats what I usually do in maya to fix this type of issue) and that has worked in the past but now it is crashing again. So it is something in the prefs The issue is that I am not nearly as familiar with the prefs in houdini as maya and I have an uncountable amount of presets, permanant defaults for almost every node, tons of hotkeys, custom shelves, galleries, toolsets, custom HDA's, custom python menus you name it. I know a lot of this is stored in the houdini install directory but does anyone have any insight as to what preferences might be the most likely culprit and how to restore the ones I need? I'm not really sure what each file does I can't go back versions either, they are all now corrupted, so it makes sense that it would be something buried in the preferences I just have way too much stuff to just trash them all Thanks guys
  13. Thanks for the info, didn't know you could control the Op_Id element, I have basically ignored it up until this point I went ahead with the object_ID vector4 variable approach which works pretty well, one thing that I didn't even think to do was just sticking that parameter in the default mantra surface/principled shader and then saving the definitition thus making it export that attribute by default without having to unlock the shader every time
  14. Is there any way to create object-based mantra AOVs (for object ID mattes) instead of having to go through the shader, like how vray for maya handles it where you can create an object property with an override object ID and you can throw multiple objects in it that will all inherit the same object ID? I'm referring to would would be object RGB object mattes in other software I know you can export custom AOVs inside a shader but for my workflow I have three issues with that- 1) if you are doing quick look-dev and iterating through multiple different shaders and you export that variable in one but quickly shift gears to a different look and forgot to include it in the new shader it won't render 2) If you want multiple objects to share the same object ID matte rather than having to annoyingly shuffle copy everything together in nuke you have to duplicate the same variables for each shader 3) and most importantly, and I don't really see a way around this, you need to unlock every shader HDA which is, as I have read, extremely innefficient in H15 as their load time and memory footprint are greatly optimized in their VEX HDA definition state Ideally this would be something you could implement at the SOP level in a wrangle or something Is there any way to quickly enable/disable all extra image planes like you can in maya/max/cinema/every other software, without manually unchecking every box? Then finally, when you export a float AOV and bring it in to nuke it basically reads it as an alpha channel; it won't be visible to a shuffle node as most AOVs would, you need to use a copy node to extract it. That isn't a huge deal except for I use a lot of python scripts that automate a lot of this and that breaks my workflow a bit. Any way around this? Any insight on this is greatly appreciated!
  15. The way I usually handle this situation is to separate all the UVs per material into their own UV tiles. Say you have an object to shatter that has existing UVs. you keep all the original UVs in the 0,0 uv tile, then create UVs for the inside shattered pieces and translate all those UVs over to their own UV shell. Do that for each material on the object, just make sure all the UV shells per material are in their own tile. When you export alembic back to maya make sure you are exporting the UVs as a vertex attribute. Here is a really great tutorial from toadstorm on seamless maya <> houdini workflow - http://www.toadstorm.com/blog/?p=435 Then to set up multi-tile UVs in maya without having to use the UDIM preset on the file node, you basically daisy chain all your textures together into the default color slot of the previous texture starting with the 0,0 uv tile. Then in each place2dtexture node uncheck wrap u and wrap v, then under translate frame, translate it over to its corresponding UV tile in the same way you did in houdini. so first texture is 0,0 second is 1,0 third 2,0 etc You can use the same trick to separate out materials too, depending on the renderer you are using. For vray you would just make black and white masks in the same way and pipe them into the mask input of a vray blend node for instance. This is also a good way to get around all the issues associated with face assignments in maya. I attached really simple examples for each of these. In the maya file I used ramps to avoid external file dependencies but you could just as easily substitute those for file nodes multiTile_uv_fracture.hipnc maya_multiTile_uvs.mb