Jump to content

spiralstair

Regular Member
  • Content Count

    269
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by spiralstair

  1. Actually the claims for RTX are still somewhat unsubstantiated. In some situations it can potentially provide big increases but as yet there has been no confirmation as to what kind of scenes will benefit. Much like the bridge between cpu and GPU rendering (think: vray) there are many tools that GPU cannot yet support, yet its lighting fast at other things. Therefore the common wisdom is that RTX technology is still very much a work in progress and it may require the next gen of nvidia cards to make the pipeline more stable. Likewise Render engines are being cagey on how or when stable releases will be available. Seemingly a 2080ti is currently a good investment, but don’t rule out the possibility of a next gen family of cards coming out sooner rather than later. Even just a scant reading of the literature shows that from high end to low end users, the Rtx family is currently on borrowed time.
  2. Ideally the motherboard should have capacity for future expansion. 1 x 2080ti is currently a better future proofed investment than several 1080ti’s. If possible a dedicated GPU for monitor, but it’s not necessarily a good idea to buy a cheap card for this. 2070’s offer the best current balance of cost for performance. For rendering the fewer higher powered GPUs offer the lowest heat and power consumption costs compared to render output, but of course your build must include a PSU which is able to support the hardware power requirements. There is a lot of hyperbole about the promise of RTX, very little of it is yet in use in stable production builds, but it’s undeniable that the potential could be harnessed within the near future to dramatically outstrip the GTX models. one thing to be aware of is that within a short space of time Nvidia tested Volta technology, for a short window the Titan V was the most powerful GPU and then the entire idea of Volta was seemingly scrapped for RTX cores. Literally within the space of a few months. And so be aware that within the product life cycle there will be a shift beyond RTX in the not too distant future. I would give it 5 years before the next major upgrade happens.
  3. Thanks @Cerbera yes that’s what I thought. Surely this is something MAXON could/should think about adding to the pipeline. There’s been many moments in the last few months where it would have been useful
  4. Hi - do we have any tools beyond motion tracker to bring video footage into the viewport in c4d? im trying to rebuild some moving objects in a video for the sake of dynamics interactions and find motion tracker does not work in all situations any workflow tips would be much appreciated (ideally without leaving c4d)
  5. unfortunately not - forester is procedural and so cannot interact dynamically with other geometry assets. If you were working the other way around you may be able to get this to work by exporting the tree/flora as alembic and then potentially c4d objects could interact with it. But in this case the tree would drive any simulations. And the alembics that forester produces are so heavy that it would take a very powerful CPU to make it work.
  6. Only that if you set / apply a material and define UV mapping before you simulate it should already be correct after the simulation. TBH I don't know if that also applies in MD but it does in C4D, say if you were using its own cloth system. CBR Marvellous Designer attributes UV's to all MD generated geometry before any simulation takes place. The UV attribute doesn't automatically transfer to your export geometry (.obj/fbx/abc) and so you need to manually select it in the export options dialogue in MD. Don't try and (re) project UV's in c4d onto MD geo - it may appear to work at first, but eventually you will see distortion especially for shell objects which contain more than one part.
  7. Hi All can anyone help me figure out a mathematical problem regarding frames per second. I am preparing an animation sequence to go onto a render farm. I have animated a character and camera - which all looks good - except that when I did a hardware render I decided the clip should be half the speed. Now I can't quite work out the best workflow for this. Normally I just leave scene FPS at at 25 and change the render FPS to 50. But the render farm I am working with cannot accept variation between these two and so I need to bake in the slow motion. With camera it is easy - just to scale the time x 2 in the animation timeline. But the character was originally made using an FK rig converted as an alembic into the sequence in order to keep the scene preparation times down (for the sake of speeding up rendering on the farm). How can I slow down an alembic? or How can I export an alembic at half the speed? so that when I bring it into a 25 FPS sequence it will effectively be playing 50 FPS? Thanks
  8. Is it possible to link a null to the moving coordinates of a specific point on an animated mesh? in fact I would like to link several hundred to different points on a moving mesh, but guidance for one would be fantastic. thanks
  9. Don't mix c4d procedurals inside a vray shader as it will cause vray to be unstable and is often the cause of crashes. It also can cause unpredictable results with DR. If you use a power shader you cannot use one of the c4d shaders inside it. Adv bitmap gives enough options to be more powerful than filter, for layers you have Tex Layer Max. There's little reason to use fresnel filter as the inbuilt fresnel options are excellent in vray. The most common limitation with vray shaders is the noise - which are far less developed than native c4d. You can in this case texture a model then bake the noise and import to vray as an adv bitmap
  10. Hi All I have an alembic animation that I have built in houdini - simple question - I would like to include several (render) instances of the same animation within a scene but each offset to start at different frames. I see that its not possible to manually modify instances - but I was wondering does instancing actually matter with alembic? since it is always referring to an external file? it's a heavy file (2GB per frame) and for the sake of rendering I am trying to keep my geometry loading times to a minimum. Is there a way around this?
  11. @Havealot perhaps you could also help me. Using your R20 file - for a separate situation - how would I go about keeping the exposure/time constant (say 50%) and then simply letting the noise animate?
  12. Hi all i have an animation sequence which includes a cloner object set to grid with 4,000 clones. The clones are static - camera moves through the clones like moving through a cloud. is it possible to randomly animate when each of these clones appears as on in render? so in frame 0 no clones appear and in frame 100 500 random clones appear and in frame 250 4,000 clones appear. By appear I mean off for render view and then on for render view next frame. once on clones remain on. im rendering in vray 3,61 and so ideally for compatibility this needs to be done in R19. I have R20 and if fields will be useful then I will move up to vray 3,7. thanks
  13. Hi I am trying to work out a way to animate a computer mouse cable. The mouse will move on a surface around x and z axis and I had drawn out a sweep spline for the cable, but am looking for best way to make the cable move in a realistic manner i.e bend a lot close to the mouse and not so much further away. I guess one way of doing it is with an IK chain. Is there a way with splines to fix some points on the spline so that the curve moves dynamically? thanks
  14. it looks like you could benefit from spending more time learning the vray physical workflow. there are several ways to learn, but the only real up to date method is to follow Stefan Laub's Vray for c4d Masterclass. It's worth every penny
  15. this is not true. Although the current implementation of projection mapping in vray for c4d is not as easy to work with as the native c4d material system. It is possible to do what you want with the Vray UV manager tag. it is by no means an ideal solution and can be a little difficult to use. But ultimately this is the most advanced way of managing multiple forms of mapping onto the same mesh
  16. Thanks @zeden I will still be going beyond January, I just need to close this particular scene by then. So if you have any time please get in touch. @Freemorpheme many thanks, I will send you a message
  17. you don't necessarily need a blend material. Depending on your setup you can use the two diffuse layers in an Advanced material - use the lower layer for the wall - top layer for decal with alpha placed into transparency slot in diffuse. You can then copy this setup for the speculars if you need different specularity for each.
  18. Hello everyone, I would like to reach out to the c4d community to ask for help for an ambitious film project that I am working on. I am an artist and academic based in London and have spent 2 years working on a 60 minute experimental film called "Quasi-Monte Carlo". The project is 3/4 completed, but I have realised that I have run out of time to complete the remaining modelling/texturing to a level that meets the work already completed. And so I am asking if anyone is interested in supporting the project by undertaking one of many small modelling tasks, which would normally take me 2-3 hours each. But there are probably 50 small modelling tasks or more and I don't have that many hours left. Here is the trailer for the film - this is not yet public - so it is a preview link for the cafe The film is entirely not for profit but will carry a visible public profile, IMDB page etc. The film initially won a highly prestigious UK award in 2017 to receive production funding from the Arts Council directly from UK tax payers. It will be shown in Museums initially in Europe, hopefully worldwide and also in film festivals. I haven't taken a fee for the work, In fact I have paid to make it and spent 2 years working on it almost to the exclusion of all other things. During the last 2 years many, many people have contributed in various ways from camera to sound to catering. @Cerbera has been incredible in providing very generous support with modelling, although it has been a lot to put on his shoulders. I was hoping that a couple of people from the cafe community would be interested in contributing towards the project, by offering a few hours of modelling time, sharing out the workload and forming a team built scene for the film. I also think it could be an exciting project to see developing here in the cafe, a way to share wireframes, renders for feedback and trouble solving. THE TASK The first major deadline is to complete a very large office scene by 5th January 2019. I have the outline geometry (below) which currently looks like a boring office, I want to rip out the stand in computers and furnish it with a lot of details, tools and machinery to make it look like the dream CGI workplace. It has 40 desks and the camera will move around looking at all of the details on the desks. We will see what is happening on the screens, but there will be no people. I have a list of core objects that I need ranging from PC workstations, to bags, clothes, office toys and I am also looking for suggestions - I can only see my own workstation - I am sure your desk is set up very differently and this would be interesting to show. I have included some images also which are of completed scenes in the film to give you a sense of the level of detailing that I am looking for. None of the models that you see are particularly complex, most of the work is in the texturing and lighting, which I can do. Is this something you would be interested in contributing towards (in the next couple of weeks?). If so I will put together a core object list ASAP and post it on this thread. There are also other things that I could use help with, I have a couple of objects to UV unwrap and a bit of help texturing (if you are a substance user) would also be amazing. Let's see what happens Many thanks Paul
  19. In the case last night the scene file was initially saved into a folder dedicated to the textured model. As in I placed the textures there initially and they did not exist anywhere else on my system. All of a sudden they are gone?
  20. Thanks @srek i see your point but I can't think of any other way this can be happening. I do a test render , all is fine. I save a backup with assets, close this file and reopen the original local file and the files have moved from my tex folder
  21. Hi i have come across a (bug?) several times recently and wondered if I am alone? intermittently when using save as with assets command, some of the texture files are moved to the new location rather than copied. for example last night I setup a render queue and found this morning that the render was incorrect as some tex files were missing. I save all files onto a backup drive, but had set the render to read the file on my desktop. eitherway, I have had this problem several times recently and it has been costly, any advice here?
  22. Hi im pulling out the little hair I have left trying to figure out which of the camera callibration tools to use and how to get the best solve. i have a source video clip, which shows an interior space with lots of useful strong architectural lines to map onto. the clip starts with a camera pan and then finishes on a still frame. My aim aim is two fold: 1) to track the material so that objects can be added into the interior space during the pan. I have solved this pretty well with camera mapper, although the scale of the camera/objects seems completely arbitrary. 2) to accurately block out the main architectural surfaces for the final (still) shot, as the stage for some dynamics collisions. The architectural modelling doesn't need to be too complex but the scale used is important as the dynamics require real world scale. on top of that the original footage was shot on a half decent canon l series lens (24mm) and i suspect there is a little bit of lens distortion going on, but when I have tried to correct this and then add it to camera callibrator the solves seem to be a bit odd and eventually off centre. So my questions here are A) can the motion camera be used instead of camera calibrator? Since the interface is far less annoying? Or or should I try and combine both methods? b) if So, with calibrator are there any rules to know about how near or far from camera lines should be set. as it happens the scene is almost square on to an architectural facade meaning there is almost no reliable z depth lines to follow. So far, I have had to solve z-depth indirectly and it leads me to weird results. i have followed the digital meat tutorial on this and watched a couple of others, but still haven't managed to create a reliable camera solve. any tips would be extremely helpful for my sanity. many many thanks p Edit - I am using a video clip for the motion camera and a still made of the last frame of this for callibrator

YOUTUBE CHANNEL:

ABOUT US:

C4D Cafe is the largest CINEMA 4D community. We provide facilities for discussion, showcasing and learning our favorite software :) Register now to gain access to all of our features. Once registered and logged in, you will be able to create topics, post replies to existing threads, get your own private messenger, post status updates, manage your profile and much more. If you need to find solution to your problem or otherwise ask for help, Cafe is the right place.
×
×
  • Create New...