paramount opened this issue on Nov 20, 2008 · 225 posts
Morkonan posted Wed, 17 December 2008 at 5:58 PM
Quote - > Quote - - Your point about the transparency detail being available on the materials for the bodysuit is well taken. I suppose such transparency could be put in as a map using an offset of the original map for the material. So, basically, a transparency map that is just a reverse of the bump map.
Somehow I have lost you. Equal if procedural material or a texture map it stretches directly on the mesh. So having transparency means the mesh becomes transparent. In this case using a second skin on V4 you will make V4 itself transparent. This might be good for a render of the ghostly hour but only there.
What I mean was using a transparency to block out a portion of the procedural texture. Then, as you correctly point out, you'd have to put in another procedural (or texture map) to make up for that on the skin of the model.
Quote - In the same way a bump is only a vision of shadow and light on same level of the mesh. On otherhand the displacement map changes the levels depending to it's texture. With high displacement you might distort the mesh. So there are no different layers for bump, displacement or transparency. They all work directly to the mesh. They have no offset between each other.
I understand. But, you can affect how those are expressed in certain areas using processing nodes before they are applied to the mesh itself. At least, as far as I know. I'm relatively new to this but, you can extensively modify the procedurals before they're ever applied to the mesh itself in the rendering process. That is what I mean - You apply transparency nodes to isolate the effects you want in certain regions before those materials are linked into the channel. So, the main body texture could have several procedurals hooked into it with different transparency nodes blended in and the transparency channel isn't touched. (My apologies for not knowing the proper terms here. That creates some difficulty in explaining my meaning at times. :) )
Quote - Technical ok, but a bit the same as above. With the bumpmapped material a bump will do it. If using the displacement that seam should be displaced. And than it could produce well visible distortions where it raises above the normal hand mesh level. A other point is that the forearm group is relatively long. So better that cuff could be placed at the end of the forearm than at the hand. But even there I would suggest to have the cuffs a bit higher towards elbows. Having them directly at the border between hand and forearm means they are lying directly in bend/twist zones between that 2 limbs. This will mess up in certain poses and/or morphing's of the character parts.
Good point. In a displacement map, they would definitely run the risk of distorting the mesh if not enough room was left for movement of the limbs. For a bump map, it may not be that big of a deal. Then again, such a detail in the texture is really only good for closeups anyway and, for that, you'd want displacement if at all possible. So, it would be best to plan on using displacement to begin with for that kind of detail and just let the user reduce it to a bump-map if they desire.
Quote - Oh, sorry, that was a misunderstanding. U-boats has been much much smaller vessels as for example today's nuclear submarines. They only used rubber dinghi's if ever neccessary.
True. I was a big curious as to why I though you meant a wooden ship's boat as being with a U-Boat. It would be extremely hard to get in and out of the submarine even if they tried to load it in the torpedo loading bay. :)
Quote - Boarding normally is done alongside the mole or dock. But rendering a rubber boat is not very impressive. So in my opinion not worth the rendering effort. Especially cause that rubber is normally black or dark grey, the u-boat hull behind is grey, the water around (harbour !!) is muddy ... darkbrown and the people are wearing grey cloth ... so a very unique grey symphony ... ok, might be with a outstanding white point ... the captain's cap but only starting the journey, not at the end ... :lol: No, that's to dark for me. Doing this with some buildings as environment give the possiblity to use some more colors.
True. If you were doing a night scene though, that black, inky image of the rubber dinghy against the moon's reflection on the water might look pretty good though. If you had crew, crouched down in the boat, and some reflections here and there, you could make it look ominous or even desperate, depending on how you presented it. I'm no art-critic nor am I a professional artist though.
Quote - I'm with you in point of the caching or better the garbadge collection of that memory. Not the menu entry "reload texture" is needed mainly but there should be a entry called "free up all texture memory". In my experience poser holds all textures as long as possible. So if I try out some different clothings with different textures while building a scene, it's best to save the result and restart poser completely to get rid of all unneccessary things first before rendering.
I agree completely. One thing I do is use a separate program to free up my memory cache on the fly while working in Poser. Luckily, Poser doesn't seem to have a problem there and will reload textures on its own during rendering if it can't find them already in memory. However, over multiple renders this causes a problem sometimes and it's best to restart Poser and let it figure out you've been "erasing its mind" the whole time. :) I'd rather have Poser force a reload of textures when I am actively using it (during Rendering) than have it sit there and hold onto all that memory just to save some texture-loading time during rendering while I am trying to work with it.
Quote - Also the filtering default setting "quality" is normally bad. The processor has lot's to do and the outcome is only blurry, worst than 50% jpeg compression. So only with hair that could be fine. For all other textures set filtering to "None".
I agree.
Quote - So surely apps like C4D have better ressource management but comparing to the prize and the number of companies having had their hands on poser, it's really, really not too bad.
I agree there as well. Poser is a really good tool set for what it is designed to do. There are other rendering packages out there that are much better but, Poser is just fine for most hobbyists and many artists.
Quote - Programming graphical programms isn't the easiest. Databases, text editor and such stuff is no big matter. But for example up today microsoft is trying to do something with graphical programming and the outcome is normally somewhat between useless and rubbish including vista's desktop.
I wouldn't know about Vista... I don't ever intend on using that. I use XP-Pro and will use that until the next Micro$oft O$ comes out. :)
Quote - So the approach of DOS/W98 was not really bad as also seen with linux/unix ...
I agree. There's a lot of unnecessary, bloated junk that comes with Window's OSs these days. I want Micro$oft to make operating systems but would rather someone else make all the front-end gadgets.
Quote - - In very complex scenes with lots of objects and textures I have to reduce the preview and go without full texture previewing except for fine-tuning.
Quote - So if you are not using the pc for gaming it's a super idea to buy a top card from one or even two generations back. So we talk about card's round 1 year aged. But opposite to the newest generation also these older top level card's are rather cheap AND they are doing their job in best manner.
Good points about OGL. I also don't tend to buy the newest cards on the market but, instead, by one or two generations behind. There's a huge difference in cost-savings and performance differences aren't always big enough to justify the extra expense, especially if I'm upgrading from a much older card.
I desperately need to get a new system. It's not money that is the issue, it's the time I feel is necessary in researching various components. So, it may take me a month or so to figure out exactly what I want, thirty minutes to order it and four hours to assemble it.. But, it's a task I really need to undertake because the system I'm using now is several years old.
Anyway, getting back on topic:
I'm still working on the project, still polishing it up and making sure everything works right. When I get ready to post it, I'll probably PM you and the other's who have shown interest here so far with a link to a "beta" version so you can test it on your machine before I post it at ShareCG, if that's ok with you. No obligation to test, of course.