Jump to content

esmall

Regular Member
  • Content Count

    235
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by esmall

  1. a workaround would be to get an external GPU (eGPU). https://www.macworld.co.uk/how-to/mac/egpu-mac-3690094/
  2. try using 3D - Spherical gradient. This uses as a spherical gradient in world space. You can set the world space in the corresponding Start coordinates. Alternatively, you could try using a Proximal inside of a Distort Shader.
  3. isntead of xpresso, try just using a Selection Object. Select all of your SDS objects, then go to Select > Selection Filter > Create Selection Object. This creates a null object w an Object tab. The tab has a list of all your SDS objects that were selected at the time of creation, and a button for "Restore Selection." You can press the button to get all your object selected, then in the Basic tab, check/uncheck Enable. Alternatively, clicking the closed eyeball in the upper right of the obj mgr will open a window that lets you select objects, tags, layers, etc. Here you could right click the SDS object, and choose Select All, and perform the same enable/disable function as in the previous paragraph. Eric
  4. I'd try using a linear cloner of cubes touching edge to edge and apply a transparent material with refraction. Then I'd fiddle with the Fillet radius of the cube, and the refraction value of the material. see attached (the camera mapped image of an apple is NOT mine, it is used for reference, and was located using google image search). hope this helps. reededGlass.c4d
  5. The surface deformer calculates based on the object's UV layout. Your object's UV's are not properly laid out. To get the MoText to project, you'll have to either current state to object (to polygonal object), or do each letter separate. Since MoText is procedural, it does not have UVs. Because of this, I believe each letter is assigned the same UV space, so it will project, but all the letters are stack on top of each other. Looks like you are using 17. I saved this in 19. In case you can't open, I took a screen shot of the UVs and a screen shot of the viewport. Hope this helps project_test_UVs.c4d
  6. the potential issue with doing that is if your cube's position is animated, the cube's axis will still remain animated in that location, but the actual cube's location may differ. See attached: the axis of both cubes are animated in the same location, but the cube's are in different places. axisLocation.c4d
  7. It's hard to tell without a test file, or knowing what you're animating (figure? object? complicated? simple?), but most likely, no. Animated properties such as Position, Scale, or Rotation (PSR) is oriented on the object's axis. Changing the axis would result in altering the animation. One possible solution would be to create a null object, and set a keyframe in all PSR properties you have animated. In the timeline, copy the keyframes from the original object, and paste them to the null object (this copy/paste is done in the timeline's Edit menu). Now delete the keyframes from the original object, put the original object as a child of the null that just received the pasted keyframes. From there, you'll have the ability to adjust the original object's axis, tweak its position, etc. See the attached c4d file. animationTransferred.c4d
  8. It's hard to tell without a scene file, or seeing an image of what you're trying to model (it looks like a flange? Or cap of sorts?) But looking at your axis, it appears that your cloner is not in the center, or hub, from where you want the cloned objects evenly spaced. Try using a Radial Cloner centered at the middle of the circle (see screenshot with blue arrow), and adjust the start/end angle of the radial cloner.
  9. Echoing Ninjad's assessment of Lens Blur, it is a hotmess. If you really need it, go ahead and use it. Another downfall of it, it is a single core effect. This means your entire comp will only render w 1 core of your computer, regardless of how many cores you have. Even if this effect is applied to 10 frames of a 1000 frame render, all 1000 frame will only render w 1 core. If you can spare the extra money, invest in Frischluft's Lenscare plugin. It does a GREAT job of applying DoF. This plugin does a very convincing job of blurring edges of high contrast/large distance objects. But yes, to get truly accurate blurring, you'd need to apply it in 3d at render time. Unfortunately, this limits the flexibility in post. There's always a tradeoff.
  10. if you have it rigged to your liking, you'll want to select all rig components and any mesh you want to make symmetrical, and go to Character > Mirror. You'll prob need to fiddle w the settings and/or watch some tuts on how to properly use this tool, but this is what you want. A symmetry object will duplicate what you have on the other side, so you won't be able to animate each side independently. Mirror function in the Character menu will duplicate the actual object(s), tag(s), weighting, etc.
  11. hmm, not sure what's up on your end, but I got it to work with the 2 images you provided. See my comp and screen shot of effect settings, see if something looks different to you.
  12. Alternative to the Tracer (a mighty fine suggestion) an interesting but little known function in the timeline might help you (if your object is freely keyframed in space): Select your object in the Dope Sheet (not position track), go to Functions (in timeline menu), and select the command Position Track to Spline. This will do what its name suggests and make a spline based on the object origin's movement through space. This would leave you to hand animate (or setup some xpresso) to animate the end/start growth of the sweep nurb. but hey, at least it'd give you an exact spline based on the object's path through space!
  13. thanks for the confirmation. I emailed them on Mar 3 regarding the issue, sent files/screenshots, etc. And I just followed up this morning. I figure if I continually follow up with MAXON, 1 of the following 2 sayings will become relevant: - the squeaky wheel gets the grease - the tallest blade of grass gets clipped first I hope for the former!!
  14. Has anyone experienced this issue and/or found a solution? In R19, when using any form of multipass, I end up with additional RGB and Alpha passes. I render as PNG with a straight alpha, therefore I do not need either of these passes. If rendering a "Regular Image" only, these passes are not generated. If rendering any additional multipasses, these passes are created, despite not being added as passes. See attached screenshots: The picture viewer is showing the appropriate passes: Background, Alpha (this is always shown in pic viewer, but never saved as an actual file w PNG images), Object Buffers 1 & 2, and Depth. In the finder, you can see the actual images generated: Background/beauty image, obj buff 1/2, depth, but also there are the unwanted rgb and alpha images. Am I missing a setting somewhere to disable this pass generation? Or is this a bug? This issue is plaguing my entire studio, so it's not just me. And I have seen posts on other forums asking about solutions. I've attached a c4d file that demonstrates this. Thanks in advance. extraPasses.c4d
  15. I would use a Shader effector, then a material that uses a MoGraph Color Shader. In the attached file, I made a separate material and used the color shader in the alpha, so I could control colors in separate materials. But you could just as easily use a single material by defining the base cube color in the cloner itself, and the changing color in the Shader effector's Shading parameter, and put a MoGraph Color Shader in the color channel of the cube's material. The actual change in color is controlled by the width of the Shader falloff. cubeRotateColorChange - mograph.c4d
  16. how about a plain effector set to rotation mode? See attached. cubeRotate - mograph.c4d
  17. not a problem. good luck! My workstation at work is a 2013 trashcan mac pro, 8 core, 3.0 ghz, amd fire 500 x2 32gb, cinebench scores around 1200. my studio is all mac presently, but I'm working on changing that. we recently upgraded an animator to a new PC w a i7-8700 CPU, GTX1080Ti, 1TB m.2 SSD. Its cinebench score was over 1400. My laptop is a 2017 17" HP Omen: i7-7700hq, 16gb ram, 1050Ti, 512gb m.2 SSD/1TB HDD. My fav part of the laptop, aside from being able upgrade RAM, HDD, and SSD: REMOVEABLE BATTERY!!!! (15" is not removable) Lastly, I have a desktop at home that I built to be a Hackintosh back in 2015. After almost 3 glorious years, I gave up and put Windows 10 on it, and have not looked back. Its older, i7-4930 (overclocked), 24gb ram, gtx960, 512gb ssd, but despite that, it still performs excellent. It still puts up ~1050 on cinebench
  18. So it looks like the XPS will offer cpu's as i5 8300, i7 8750, and i9 (not sure of model). Direct answer to your question: yes, the higher the number the better the performance, with caveats. The different levels of the Intel chips can be thought of this way: i3 - entry level - web browsing, maybe some word processing i5 - moderate user - word processing, spreadsheets, photo editing i7 - professional user - 3d animation, video editing i9 - ludacris mode (Spaceballs reference!) - for the 3d artist in a hurry Based on the numbers, the new i5's are competing on the same level as the i7's were a couple years ago. Someone with an actual ComSci background could probably give you more details as to WHAT is different and WHY the i7's (and subsequently, i9's) are better, but that is beyond my knowledge. Here's the specs of the known CPU options: i5 8300 - 4 cores, thread out to 8, at 2.3 ghz, boost to 4.0 ghz i7 8750 - 6 cores, thread out to 12, at 2.2 ghz boost to 4.1 ghz Boost speeds are essentially the CPU being overclocked on the fly. Caveats explained: So the i5 has SLIGHTLY higher frequency at base frequency, but this will only matter when performing single core operations (which is a fair amount of operations in C4D). However, this is such a minute difference, I would opt for the i7, as it gains you a slightly higher boosted speed, plus you get an additional 2 cores, that thread out to an additional 4 to help w render power. Regarding your GPU question: this only comes into play if you plan on using ProRender (which I'd advise against, it for now), or a 3rd party renderer such as Octane, Arnold, etc. True, the viewport in OpenGL mode uses the graphics card, but it will not remotely tax a the 1050Ti GPU. That being said, my laptop has a 1050Ti, and it works fine. Nothing spectacular, but it gets me by. I do not use it for 3d rendering, but I use its CUDA cores to make After Effects perform better. All this jibberish aside: the new Dell looks great. But, as with every hardware purchase, that's If you can justify its price. By all means, go for the upgraded CPU, if you can afford it. Do you work with C4D for a living? Or is it a hobby? Are you using the computer daily? How long are you hoping to have the laptop as your workhorse? When was the last time you bought a new piece of hardware? Have you gotten your investment out of your previous purchase(s)? The budget is a personal questions that can't be answered by a spec sheet. Thanks for that blog post spec sheet. It's good to have a written record. Most of my knowledge has been gleaned by the MAXON reps and industry experts over the years while talking shop at various conventions!
  19. I think some clever modeling, either with some extrudes to give the appearance of plates that make up the lips, or separate/discrete geometry. Either way, I think the pose morph would be your friend here. See the attached files. One would be a setup using a continuous mesh, the other would be using separate geometry. Both setups are using the posemorph tag. robotMouth.c4d robotMouth 2.c4d
  20. To echo Cerbera's comment, I would approaching this as a mechanical rig. Look up some tutorials on this topic (search for robotic rig, mechanical rig, etc. Another suggestion: go to a hardware store and pickup an actual SOSS hinge in order to appreciate the intricacies of such a hinge. Being able to put hands on such a piece of hardware will reveal a lot of the fine details. Or at the very least, find a good video of a hinge in action, and watch it frame by frame, forwards and backwards. E
  21. hey Bass, If you're using a built in render engine w C4D (Pro Render aside), you're talking about CPU power. The more cores and the greatest frequency will be your best bet, but you'll need to find a balance. If this is going to be your daily driver, bear in mind that the majority of tools in C4D are single core (exceptions existing, of course, such as the hair module which are threaded). For single core operations, you'll want to get the highest core frequency you can get, regardless of core count. However, when it comes to render times, core count will trump frequency. Unless you're planning on using Pro Render or a 3rd party renderer, the GPU doesn't make much difference (obviously other applications will differ; After Effects for example). I wouldn't go out of your way to get 64gb of ram. Ram will not improve render times or viewport performance. Both Physical and Standard renderers in C4D use the CPU, not even GPU. So make sure the CPU is where you're spending your money. If 64gb doesn't break the bank, go for it. Otherwise 32gb will be perfectly fine. (Personally, I have 16gb and I work just fine.) Looking at what's online, I see some PC laptops that are using the i7-8750 which would be a solid bet: 2.2ghz boost to 4.1 with 6 cores that thread out to 12. Personally I have a laptop w the i7-7700HQ (2.8 boost to 3.8 w 4 cores threading out to 8) that I bought about a year ago, and it works great. You are correct that a desktop is a better bet for rendering. However, with the availability of online render farms now a days, you can easily get away with a laptop as your daily driver. And you always have the option to hardwire some computers together to make a mini render farm.
  22. sounds like your xpresso setup is using Absolute nodes instead of Relative nodes, to reference the path of the object from which you're trying to create a trail Question for you: Why not just use the MoGraph Tracer object, then Sweet the Tracer? I think I've seen some Xpresso setups online to create trails from objects' paths, but usually those are done for people who don't have C4D Studio version (lacking MoGraph). **Edit** by Sweet the Tracer, of course, I mean, SWEEP the tracer
  23. to the best of my knowledge, the answer to your question is "no." And rightfully so in my opinion. Imagine how confusing that would become if you opened a file that had been setup that way a year from now! Or if someone else had to dissect your file. Having referenced or instance nodes in xpresso setups would get very confusing very quick. Like you said, I think your best bet is to create some user data as your "bridge" (that's a great term for it, by the way!) Whether you do it this way, or you figure out a way to create your instanced node, make sure your priorities are correct!! Tags are calculated left to right, top to bottom. Having them out of order will cause calculations to be computed out of order, leading to rig lag. You can always tweak the priorities in the Xpresso tag's Basic Properties if you run into issues.
  24. Animating sutures accurately/believably is no small feat. I see you're using C4D Lite in your bio, is that accurate? That may seriously limit you. What kind of suture are you doing? (the attached image is not mine, but from a simple google image search for suture reference) Long story short: this shot can be as easy or complex as you want to make it. You could draw out your final spline, including the knot, and animate the End of the Sweep Nurb. At the opposite end of the spectrum, you could include hand animated tissue dynamics of the thread passing through the skin, thread dynamics using something like the PointAutoRig.py script (look for that from CuriousAnimal). Or you could go anywhere in between. Good luck!
  25. If you've paid for MAXON's MSA, you have a free membership to Cineversity. Just email MAXON and ask for your coupon code. Otherwise, without some coding, you won't be able to render 6 cameras at one time. If this is for a paying job, and you don't have the MSA/free Cinversity membership, I would think this might be worth the expense of a Cineversity membership. Or justification for upgrading to R19. Work smarter not harder!

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...