seedub

Thanks, I've sent the file to Otoy to see if they can help solve the issue.

So setting all Hair settings to Generate > Circle seems to solve the Live Viewer issue now, but in turn it removes all colour from the hairs. Any pointers as to why this is happening and how to fix it would be great!

I have 40+ bodies inside a cloner, all with varying hairstyles on. The cloner is set to linear and only showing 1 body at a time. In my Octane Live Viewer I can see the 1 selected body but also every hair module from all the other bodies are showing at once. I'm unsure whether this would be a cloner issue, an Octane problem, or something to do with the render options of the hair. I have tried changing the 'Generate' settings on the hair module, I've tried applying an Octane tag to each hair module, tried changing settings on the cloner itself. All to no avail! Any help appreciated, before I pull all my own hair out. Ta!

This is great! Thank you very much. I'm struggling to find a way to keep the random integer constant though, so I can manually create a new seed without the 10F limit. Is there a way I can create a button in the viewport that could re-generate a random number when pressed?

I'm looking for help in building an Xpresso module that would randomise the offset of multiple cloners. I'll have 10 linear cloners all containing various character parts, with the offset selecting the parts. I'd like to be able to generate a new character combo with 1-click to test out some random variations. I assume I need to set a range and feed in a random number to each cloner's offset value, but I can't seem to find the right node order for Xpresso. A point in the direction or a quick breakdown of the Xpresso would be great. Ta!

I have a stadium & pitch modelled in C4D and imported in to AE, where I have animated cameras around 2D character motion. The motion is all matched up with the pitch dimensions in AE, yet when I bring these cameras back into C4D to render the background & environment, the cameras no longer match up with those from AE. The cameras seem to be tighter on the focal length, or on a slightly different path. Are there any known issues or workarounds for matching up the cameras properly? Or am I missing something obvious? Or should I just admit defeat and do all the cameras in C4D using placeholders and re-do the motion in AE....shit. But any fixes to the above appreciated!

Thanks for your reply, we localised all projects & assets to C drive and still the same. We just compared every C4D/Octane setting and it seems to have been the Out-Of-Core settings were too high. I brought the RAM usage and GPU head room down to 10GB and 1024MB and my 14min frames are now 57s. That'll do for now! Thanks

We have 2 pretty much identical machines here with vastly different performance in terms of navigation & rendering. RTX 3080 & GTX 1080i 64GB RAM AMD 3960X 3.79GHz C4D R23 & Octane 20201.5-R3 Opening the same project from the same shared drive, my colleague's machine exports a frame in 43 secs vs my 2:20 min. Another project we're working on, he gets 1.5 min vs 14 min on mine. I just noticed a note in the Octane log: device 0: disabling RTX acceleration as mesh data cannot be placed into VRAM I haven't been able to find much about this. Could anyone advise on where to look and any possible fixes for this particular error, so we can eliminate this before troubleshooting elsewhere? Thanks

Yeah I assume when I open it at home and the Mac tells me 'these Octane plugins are all missing', it must be disabling a lot of stuff and I guess I want to be able to replicate that at the office, by 'dumbing down' the project to its bare-bones. This is what I tried to do by disabling materials and shading. But your comments on the cloud are probably the main issue, it could be a pCloud/Win issue. I will run some tests from portable SSDs and see what I find. Thanks for your help.

No, all projects are being tested from the cloud, whether at home or in the office. I will localise this particular project and try it from an external SSD this evening on both machines. And I find the same issue regardless of Octane, if I disable all scatters and delete all materials on the PC and just work in the viewport with wireframe I have the same issues. When I first discovered this performance difference, I was working on a character rig without any Octane elements and the difference in viewport wireframe rendering was staggering. So Octane seems to be irrelevant in this problem. Yes they are 'unrelated' engines but surely the PC should outperform the Macs when opening the same project. They are comparable in the sense that the 7 year old one is working better than the 6 month old one, and I was wondering if I'm missing something obvious in setting up the PC for C4D viewport performance.

Hi, yes the original posts are regarding Octane on the office PC, and non-Octane at home on the Macs. The files are all on our pCloud storage, but I think it works the same on local drives too, I'll double check that now! Yes, I see Nvidia Control Panel as an option.

I'm back, 6 months later and still struggling to figure this out. I'm working on a project involving 6 bird rigs, in a fairly basic Octane environment. I work all day on my RTX3080 machine at the office, then I take the project home and work on my 2014 iMac or my 2015 MacBook Pro, and they both outperform the PC in terms of viewport navigation, timeline scrolling, moving assets, animating the birds (including dynamic hair/feathers) and especially viewport rendering. On the PC, I can disable all materials from the viewport, set detail to low, set display to Hidden Line, and I still can't scrub the timeline or jump to a particular frame without a huge lag. On the iMac it's almost like scrubbing through a finished film, I can see every detail on every frame as I shuttle back and forth. How can one project be interpreted so differently on a brand new machine VS a couple of haggered old Apple machines? Anyone else had similar?

I have a flock of birds to style and dress, and not a lot of time to do it... I'm trying to establish a hair coverage for one, before implementing it across the others. I'm having very mixed results with the coverage though. See attached. Some of the hairs insist on growing inside the surface. All normals have been aligned/reversed/optimised to no avail. I've tried setting minimum distances between the guides, and the hairs. All guides and hairs are set to polygon centre. Any help appreciated! (please move post if in the wrong thread)

Solved: Had to convert the cloner to objects. This fixed the offset problem, which allowed me to disable the connect and resolve mesh & lighting issues.

Thanks Deck. They are following some curves and sharp corners unfortunately, so it seems Spline Wrap is my only option for the dog's movement at the moment. All spline wrapped objects are held inside a radial cloner too, it seems it may be the cloner which is messing up the offset paths. If I disable the cloner, everything works fine. (But the Cloner is required to complete the 125f loop) If I enable the cloner, I need the Connect object to keep it all flowing smoothly. The Connect object causes all my materials, mesh surfaces and lighting to go weird. Is there any way to bake or cache the spline wrap's cloned motion?

Hi, I have a multitude of hot dogs flying along splines using the Spline Wrap deformer, and they're all then inside a cloner. The animation of the spline position became very jilted, and the dogs would pause at seemingly random moments, whilst other dogs were flying along happily. Someone suggested placing all the content in a Connector with 'weld' turned off, this worked perfectly at first. Since doing this my lighting, materials and edge smoothness all seems to be affected. Granted, this is the first time I've used the Connector so I'm unsure as to what it's actually doing. It's causing more problems than it resolved, so is there an easier fix to the spline issue than using connector? Or any way to tame Connector? It looks like splines are working fine in a hardware render, but not working in an Octane render. Thanks C4D R23 & Octane

Thanks Srek, here's the results. Any other results I should be exporting or is this enough?

Cinebench results attached. Ryzen Master results coming. I just noticed if I let the viewport render keep running, after 50 frames it drops to 1s a frame, then starts rising again after a few hundred frames.

Thanks for the advice, I'll try all of this tomorrow in the office. I'll be back with some results. The PC was built by PCSpecialist. Just checked and the RAM is Vengeance 2400MHz... not ideal then.

I was trying to where possible, but I planned on using the road parts along a Spline Wrap in the final scenes. From the tests I did, the render instances weren't working when placed inside a Spline Wrap.

I'm trying to avoid a 'why's my C4D so slow' post here, but also very much one of those posts. I've installed R23 on a brand new PC build - 64GB RAM, RTX 3080, AMD 3960X - and definitely not seeing the render times I expected. I'm working on a project on 2 different machines, the new PC and a 2014 iMac 5K 27". I'm seeing much better performance on the iMac than the PC. iMac - 32GB RAM, i7 Quad, AMD Radeon R9. The scene contains a ToonRig character, a low-poly bike, and a low-poly scene built mostly from instances. For general viewport navigation I have to disable pretty much everything and work in Box mode with quality set to Low, even just to move around the scene. Barely even started animating the character yet. Character is 10K polys. Whole scene is 21K polys. Speed Tests: Hardware / Viewport Render (only character enabled): PC: 10s iMac: 1s Standard Render: PC: 3m iMac: 7s Hardware / Viewport Render (full scene): PC: 11-25s iMac: 0-5s These exports go up to their slowest times then all drop to the quickest after around 50 frames (what's the reason for this too?) I guess firstly, is there anything obvious I may have missed during set-up after moving over to PC? Can't be anything that different between PC & Mac set-ups to unlock the extra power?? Secondly, what would be slowing such a simple scene down? Surely it's the character rig, but it's a fairly basic character so I'm unsure how best to work with it, other than in Low and Isoparms Mode. Thirdly, Is there any kind of scene test file to ensure everything is running as it should on the new build, speed and compatibility wise? Any advice appreciated. Scene file available here - https://we.tl/t-USSU5ZVoED

I'm rigging a bike for a character test and looking for a solution for some automated pedal rotations. The main middle/crank shaft is rotating, the pedal arms inside that null also spinning, and the pedals themselves are parented to another null to keep them flat whilst in motion. I'd like to add some input data to that static null so whilst the pedals are being pushed down they are tilted backwards 45°, and on the up stroke they'd be tilted forwards 45°. Is linking the middle crank's rotation to the pedal's control null with the Range Mapper the best option? I've tried playing around but to very mixed results!

I've built a character using zBrush models with multiple instances controlled by a pose morph tag. I'm trying to animate them using squash, bend and twist deformers. All the separate elements of the model (hair, glasses, eyes) deform well together, but the body deforms at a different rate to the others. The body is the only element with the pose morph tags on, morphing between varying mouth movements. Is there a hierarchy issue to get these working better together, or do deformers and pose morphed models not really work consistently?

I'm working on a zBrush character and exporting the body elements as .OBJ to use as pose morph instances in C4D. I'm looking for a way to 'link' or point to the obj in the project folder, which would ideally update everytime a change is made in zBrush and exported again. Is there any utility in C4D to allow this? I'm looking at XREFs once the whole character is complete, but I would like something to avoid having to scale and re-position the body elements on import in the meantime. Also looking at GoZ at the moment, having some trouble installing it but I assume it would also require re-scaling/positioning each time it updates.

Thanks, that worked great!