Jump to content

Cerbera

Moderator
  • Content Count

    9,811
  • Joined

  • Last visited

  • Days Won

    297

Cerbera last won the day on March 20

Cerbera had the most liked content!

Community Reputation

2,111 C4D Cafe High Order Member

About Cerbera

  • Rank
    Imperial Archon
  • Birthday 02/28/1973

Profile Information

  • First Name
    Jay
  • Last Name
    Wood
  • C4D Ver
    20.057 Studio
  • Location
    South East, UK
  • Interests
    Aerial Photography, Pianoing, Guitaring, Bassing, Drumming, Audio Production, stupidly fast Electric Unicycles, String Sections, Flying machines, 3D things, technology, critical thinking, cats, nature, skies, large hairy spiders, robots, dark scary synth drones and Omnisphere.

Recent Profile Visitors

15,415 profile views
  1. So what are we trying to make here ? Standard wool carpet ? Or some other type ? Have you got any reference photos that show the sort of thing we're aiming for ? If it is the woollen type, then I think the mistake we're making is to use one thick 'hair' per clump, rather than lots of smaller hairs clumped into your larger groups, which i think will just give a more carpet-like effect. I'll have a play with that later when I can see what we're trying to match... CBR
  2. I have been doing further tests. Not finding anything I'd call conclusive, but I would say it's there even with GI, so I think the problem is in the hair somewhere. I think it's something to do with the sheer amount of hairs and the fact that so many of them intersect each other. If we turn down the end thickness of the hair to something like 0.2 the problem disappears because (if my theory is right) nothing is intersecting there anymore. Don't think I'm seeing any banding in this ? So, that seems to be one solution, although of course it may not be the one you want. The other way to address this (again, if my theory holds water) would be to lessen the amount of hairs, whilst keeping start and end thickness as you had them. Here I culled about half the hairs you had going on in all 3 types... Again, not seeing the banding so that seems to have also worked. So, that's my evidence - don't know how correct the theory is, but at least you have things to try... Let us know how you get on, upload a result you are fairly happy with, then we can work on adding realism... CBR
  3. OK, while I can see what you were going for there, in keeping your edge flow horizontally suitable like that, we do have to compromise that slightly to get enough geo for the buttons. But rather than getting that curve early on, and then manfully struggling later to get holes that do not disrupt that curvature, I think I'd start flat, get symmetry out, and start outlining those holes with the minimum amount of topology you need, which I judged to be this... We need to keep our rims thin when they are this proximal. When connected, symmetried, and SDS'd, that looks like this... ...which I decided didn't have enough topology under the biggest circle, so changed that thusly, and then got my FFD out to match the curve of that section... Notice how the edge flow still goes mainly straight across the panel in the right direction to support the curve, but now does so in a way that also supports the holes. It's a little higher poly than I would normally like, so perhaps were I to start the body after this, I would do it unconnected initially, by extruding out once, then splitting off that outer ring and deleting every other edge around it, and continue at 'half-res' until I had the basic shape, then apply 1 level of SDS to restore the subdivision we had before, lastly connecting it up to the buttons section. Does that make sense ? Of course it won't be the only way to go, but it is what I'd try first... CBR
  4. Very frustrating that I have to go out now :) If you can wait an hour or so, I can very much help with this when I get back... CBR
  5. Interesting. It's only there when shadows are turned on with your lights. I tried all sorts of different settings there, and still the banding. I have to go out now, but will do further tests when I get back if the other guys don't beat me to it :) CBR
  6. We should also mention posting in the right category. I have to move 98% of the threads from new users in the cafe every time they post ! CBR
  7. I don't either - they shouldn't be marked as such. I suspect it's simply a bug (or bugs), and something has just been overlooked about that functionality since the new core updates... but if we have told MAXON, and notified Igor here at the cafe, that is all we can do. CBR
  8. You will need to upload the scene file... otherwise we're just firing shots in the dark... CBR
  9. Fancied a quick little personal weekend modelling project, so thought I'd do the Golden Snitch from Harry Potter. Painfully easy to do with either Volume builder or Sculpting, I fancied more of a challenge so I did it properly, with SDS. Here we are at the end of Day 1, most of the modelling done, but no textures yet... 100% quads, but still the odd 6 point pole remaining :) Of course it's different in every film, and you can tell they didn't know it was going to have to open until film 6, but I'm doing the film 3 version, arguably the best of the series because Alfonso Cuaron was in charge... Texturing tomorrow... CBR
  10. Yeah, had I have seen this before you did it I would have cautioned you to back up the .lib4d file rather than deleting it. The online updater keeps a record of what you download, but not what you uninstall manually, so if that option is no longer available you'll have to contact MAXON for a direct link / their advice. CBR
  11. You're welcome. But don't just take my word for it - we learn by seeing and doing, so setup your file with some volumetric lights, render a single frame once to TIF, then again to png, then to a short section of video. Compare those in the Picture viewer. If those are significantly different, upload them, with the scene file so we can see what other reasons there may be for discrepancy... but they shouldn't be that different - they should be almost identical. CBR
  12. No that shouldn't happen, and tbh I don't believe it does ! ;) Please do feel free to upload a scene file that demonstrates this. The format shouldn't change anything about the render except how much compression gets applied to it. CBR
  13. Render to png from Cinema, then whatever video format you prefer from your post application. I don't know what you have installed, so can't really advise which codec is best for you... CBR
  14. The artefacts you are showing are compression artefacts caused by rendering direct to a compressed video format or to a similarly lossy format like jpeg in the case of stills. The reason we output to png (or TIF / PSD etc) is that a) it is lossless, so this artefact problem doesn't occur, and b) if the computer crashes or there is a random power cut at frame 2999 of a 3000 frame render, then we don't have to start that render all over again because we were clever enough to save single images ! We'd just have to resume it at the point it stopped, whereas if we were rendering to a single video clip, that clip would be broken, and we'd have to start again from frame 0. When renders can take an hour a frame, that is a serious advantage, or rather you'd be mad NOT to do it.. Lastly, the vast majority of output from Cinema benefits from post processing, and all decent Post-Production apps import still image sequences, so we might as well do it as standard. Of course you can render out a lossless uncompressed raw animation, but given the above - why would you ?! CBR
  15. Lols - well strictly speaking triangles in modelling ARE bad :) But they are only bad technique - the program should still deal with them (especially outside a Sub-D workflow) with no issues at all - indeed the Bevel tool and deformers both create them with default chamfer settings (a mistake IMO). However what I do find interesting is that if I recreate that problem shape from scratch, and use the bevel deformer with 1 segment or uniform mitering (which gives me an almost identical result to yours but all quads), I do notice that doesn't produce any polys marked as errors, and indeed doesn't seem to give the shading issue ! But tempted as a Sub-D fanatic such as myself would be to use this to extol the value of an all-quad workflow, that would be wrong, because at the end of the day the mesh checker shouldn't be marking those triangles bad and phong breaks should be respected by render in all cases, and regardless of topology (except perhaps ngons). CBR

YOUTUBE CHANNEL:

ABOUT US:

C4D Cafe is the largest CINEMA 4D community. We provide facilities for discussion, showcasing and learning our favorite software :)
×
×
  • Create New...