The sci-fi crate contest
Community Forums/Graphic Chat/The sci-fi crate contest
| ||
| I am interested to see who can make the best sci-fi crate, mesh and texture. Anyone else want to participate? Polycount limit ia 2000. 1,048,576 texels are allowed, either as one 1024x1024 texture, or as several smaller textures. The deadline is Monday, August 30th. (So post it Sunday night.) The winner gets to pick an item for everyone to model next week. |
| ||
| and the prize? or is it for the honor? |
| ||
| <snip> |
| ||
| Just so that it doesn't look like there is an ulterior motive to this, why not make it so that all entries remain the property of the artist who made them? Sound fair to you? |
| ||
| Yeah, nice try Halo. :) |
| ||
| They have a point. |
| ||
| ...why not make it so that all entries remain the property of the artist who made them? Okay, everything remains property of the author. I am going to make the best one anyways. :P The prize is you get to choose the next contested item. |
| ||
| I have begun work on the Uber-crate! |
| ||
| Heh, I might as well join in if it stays mine. |
| ||
I made my mesh. It took 2 hours and 50 minutes, but I am happy with how it turned out. A couple of normals are bad because I haven't merged the sides yet, so I can UV map them all the same. I will do bump maps and textures over the weekend: |
| ||
| Not bad Halo... but I can make better (have already done for my game) :) I'll tweak the texture during the weekend and post it here then. I assume that the model has to be "game-ready", ie not a million polygon mesh? |
| ||
| Dang, Mustang wil kick my ass at this. I will post polycount and other specifications at the top. |
| ||
| 2000 Tris for a crate? This is more than my characters have. |
| ||
| For reference, I think the Unreal2 primary crate had 500 tris... |
| ||
| wait till you see my crate. it will have 6 polys |
| ||
| you could actually have a 4 poly crate, since no one said a crate has to be a cube lol. if it allways lies on the floor you can optimize it further and make it a 3 poly crate. Since often in games that are not locked to convex hull geometry, you remove underside faces and faces you can't see like those against walls etc. |
| ||
| If you make it a sprite, you can get away with 1 poly. |
| ||
| 1 is the minimum polycount. |
| ||
| Here is my crate: [|] Bloody class that - I'd like to see anyone else do a bump-mapped crate that small - notice the floating point texture stacked interlopation and reverse dynamic physics (mapped into the texture - frame by frame - every frame - every second). You can use it in your game "halo". |
| ||
| I'm not working on a sci-fi game. Okay, let's be serious. |
| ||
| Normal mapped sweetness count? |
| ||
 |
| ||
| 2000 Tris for a crate? This is more than my characters have. For reference, I think the Unreal2 primary crate had 500 tris... Current Triple-A games have easily 5000+ polygons for main characters (even DOOM3 with it's normal maps), and rendering single surface objects that have below ~2K triangles with current mid- to high-end cards just leave the card idle - it's just as fast to draw 200 polygon object than 2000 polygon one. Of course total amount of polygons (vertices) in the scene to be rendered do count because VRAM has to store those too in addition to the screen buffers and textures. But 2K crate is reasonable IMO if you don't have huge warehouse full of them. My crate has exactly 600 tris - and it has separate lid/top and full interior, is fully beveled, and has been *very well* UV-mapped. I designed this for (my) game use so I tried to shave off every extra polygon and try to make it look more complex with good texture map(s). |
| ||
| Also, the mesh data would take much less VRAM if you instanced it, so you can have a warehouse full of them and take the same VRAM as one copy. |
| ||
| But a warehouse full of crates would kill you on surface count. Ps: Who's going to be judging which crate is best? |
| ||
| surface count is a blitz invention. Engines like serious sam render 100 seperate enemies charging at you and that takes a similar hit to a few large enemies. a modern engine won't have this lousy idea of surfaces. instead you will personally set the state of the card to render and render a whole batch, then go to the next one. this is ideal for moving objects and seperate objects in game that are basically the same but with different transforms. it's a shame blitz went with the surface and brush concept as that truely sucks when you want speed. |
| ||
| "it's a shame blitz went with the surface and brush concept as that truely sucks when you want speed." It makes it easy to use, and I don't see why Blitz should not be able to render multiple entities that use the same mesh / brush automatically, in batches. I think Mark chose to render objects in Z order rather than surface order, which is what resulted in the surface slowdown even when all surfaces are painted the same, but I can't say for sure. However, it seems a possiblity given that Blitz will slow down even if you have entities which are not on the screen. I recently mailed Mark about this. I can actually make Blitz render SIGNIFICANTLY faster if I determine which entities are off the screenin Blitz and hide them manually. And that is without any collision detection. And this beats using EntityAutofade to perform the same task. Something's not right there! |
| ||
| Yes, indeed, because blitz occludes thigs that are in the back of the camera anyway. This is weird. But back to the crate thing: how about a challenge tha allows no more than 50 Polygons? Or one that gives etrx points for low polycount? Yes, I'd also like to make some dot3 bumpmapping fx on a simple cube with 12 Tris. BTW. the surface this isn't that bad. I found out that when you add all (static) meshes together to one mesh, surface redundances are removed and the number of surfaces is optimized, this way things render real fast. |
| ||
| Jfk: If you win this challengs then you can specify that for the next challenge. :-) |
| ||
| yeah I noticed similar weirdness myself. After having come from developing on Xbox and PS2 on an engine that really didn't care how many entities were in a level. we had hundreds od destructable static destructible objects with damage states. Using blitz is a bit weird as I have found that om a low spec system its actually better to ignore the fustrum culling and basically take every same surface mesh in the level and combine it into one single surface object. I rarely have more than 20 surface objects allthough I do often split large full level meshes into quadrants. There just doesn't seem to be a significant speed increase with building a level in a more logical manner particularly if your going to have a lot of interactive gameplay objects in the scene like weapons, NPC and PC meshes, FX etc these usually bloat your surface count so I end up using as few static etities as possible. THanks for the Manual culling tip, will have to get Jeremy to try it out and see what kind of bonus we get out of it. |
| ||
| Please discuss surface rendering concepts in a separate thread. It's interesting, but doesn't belong here. |
| ||
| Indeed. |
| ||
| Nobody will beat my crate anyways. |
| ||
| why not normal map these? also, should post orthogonals + perspective wirframes. perspective should have the TRIANGLE count (not the polys since some people work in quads). should also post those skins and texture maps. then one beauty render on a "storage space" floor (concrete, metal, etc). <note: i'm just regurgitating what most CGChat, CGTalk and other sites use> |
| ||
| why not normal map these? Because Blitz doesn't support them... not properly and not tangent space, so it's a no-go for games? also, should post orthogonals + perspective wirframes. perspective should have the TRIANGLE count (not the polys since some people work in quads). should also post those skins and texture maps. then one beauty render on a "storage space" floor (concrete, metal, etc). Sure, why not. Eeeezy. |
| ||
| Blitz supports them fine, you just have to work with enough verts though. |
| ||
| Blitz supports them fine, you just have to work with enough verts though. BS. :) "Supports them fine" would mean that I load a model, apply tangent space normal to it, and them just move/animate the character/vehicle anyway I like and Blitz shows the normal map automagically and correctly. Updating vertex normals "by hand" every loop and other unnecessary "fixing" etc is not what I call "supports them just fine" :P Oh, and the screenshot(s) from my compo entry will be posted closer to the deadline, I'm got carried away modeling my "Mtl.SPW.Sd.Kfz. 251/1 Ausf.D mit schwerer Wurfrahmen 40" [they couldn't think of any longer name, eh :)] |
| ||
| a crate can satisfy the requirements of several gaming conditions: - regular low poly, skinned object - bump-mapped low poly - normal-mapped low poly they could all be the the same thing. i brought it up not thinking it would be of immediate use, but rather in hoping to extend the thing's useful life. someone starting a game today may suddenly find normal-mapping a viable option next year. |
| ||
| I posted a request for a function to color a vertex according to light vector and normal, and no one could answer it. So Blitz does not support normal mapping. |
| ||
| I give you ubercrate. 812 vertices, 1048 polys. I think it looked better before I textured it. Polys  Wireframe  Textured  Download the whole thing here: http://www.leadwerks.com/post/ubercrate.zip Minor update, cut the vert count down to 600-ish, minor changes:  |
| ||
| . |
| ||
| [quote] Blitz supports them fine, you just have to work with enough verts though. BS. :) "Supports them fine" would mean that I load a model, apply tangent space normal to it, and them just move/animate the character/vehicle anyway I like and Blitz shows the normal map automagically and correctly. Updating vertex normals "by hand" every loop and other unnecessary "fixing" etc is not what I call "supports them just fine" :P [/quote] *cough ... Actualy you don't need to do any vector manipulation "by hand". You can just use a cube map containing the lighting normals with your tangent space normal map DOT3 blended atop it ... then the models can move and animate with the "bumpmap" rendering completely taken care of useing only the GFX card :) |
| ||
| *cough ... Actualy you don't need to do any vector manipulation "by hand". You can just use a cube map containing the lighting normals with your tangent space normal map DOT3 blended atop it ... then the models can move and animate with the "bumpmap" rendering completely taken care of useing only the GFX card :) Show me a working example with code, even if it's just cube with whatever tangentspace normalmaps... |
| ||
Hi-tech sci-fi crate : 822 polies |
| ||
| Show me a working example with code, even if it's just cube with whatever tangentspace normalmaps... I'm not the first to think this up and ATM - I don't have any web space to upload anything :P But you can find out more here: http://www.paulsprojects.net/tutorials/simplebump/simplebump.html and here: http://planeta.terra.com.br/educacao/alessandroborges/bump/bump2.html as well as some Google searching... The idea is very simple...instead of going in "by hand" and setting verticy colors to reflect thier normals...you reflect those colors from a cubemap (even works with a spherical map) If you are feeling experimental and have time for a quick do it yourself example then fire up Gile[s] then copy both the normal map and light map images from http://planeta.terra.com.br/educacao/alessandroborges/bump/bump2.html or use these direct links: normal map =  lightmap =  go into gile[s] or Blitz3D and create a terrain, then go to it's material..set the surface color to white (255,255,255) with fullbright...then bring up texture number 0...load in the above lightmap (the funky smooth colored one) and set it to spherical before applying it...then bring up the next available texture slot and load in the normal map from above, setting it to Dot3...and presto! You can move the camera around looking at the terrain, it almost seems like you have a bright flashlight hitting it...you can play around with the terrain surface color...changeing it to full on blue (0,0,255)brings out a lot of the detail...but note that both the above linked normal and light maps are rather low quality JPGs, which acounts for the artifacts...the lightmap could use a blurring ot two in image editing software to remove some ogf the "pops") course that is with a basic spherical map and a non-tangent space normal map...with cubemaps it works much better :) |
| ||
| MSW, the first example is OGL, the second Java - I know that normal maps work IRL because I use them at work everyday, but I need someone to convince me that you can actually use tangent space normal maps in *BLITZ3D*, easily, without any hacks. I'm still waiting. |
| ||
| Wolron, no offense, but where did all the polys go? You could make that in about 12 polys, or at least less than 50 if you worked some of the detail into the texture. |
| ||
| Beaker: Why the beautifully beveled edges of course! |
| ||
| Very nice crate. Want to upload the model? I am trying to figure out which parts are the texture and which parts are geometry. You should also post a wireframe render with something like this. MSW, I think you are giving an example that may work just fine in the very simplest case, on a sphere, but I don't think that would translate through to working on the walls of a map. The image you posted isn't even a cube map. Did you ever stop to think that someone might want lighting from a DIFFERENT direction? I know it's a wild concept. You really need to think things through before you say something like that. |
| ||
| "I am trying to figure out which parts are the texture and which parts are geometry." I'll give you a hint Halo. It's a BEVELED CUBE. |
| ||
| Those indentations on the side might be geometry, too. The polycount is way too high for a simple beveled box. 6 sides x 2 polys + 12 edges x 2 polys = 36 polys. Wolron, do you want to pick next week's model? |
| ||
| They aren't bevelled, they're rounded. That adds up -fast-. |
| ||
I added a winch attachment. I think it helps make the holding beams look a lot more functional. With that, I think this mesh is complete. I did enjoy making this:  |
| ||
| "Wolron, do you want to pick next week's model?" I thought the WINNER, got to do that? It's not August 30th yet. Also you never specified how a winner would be decided. I say we take a vote in the thread. |
| ||
| Oh, I meant August 22, but since I said that, we'll keep it open for another week. I don't like voting, because people will vote against me for no reason. It should be pretty unambiguous who makes the best model. I might win if my skinning ability didn't suck. My guess is Mustang will take this, but we'll see. |
| ||
Here is a version with just the holding frame: Ubercrate is a multipurpose shipping solution! Order yours today!  |
| ||
Mine, untextured as of yet: I expect the cage'll look something like this metal:  |
| ||
| If I can make a suggestion, use a shiny environment map on the recessed portions of the box. It looks really good when only some surfaces have cube mapping. |
| ||
| Well, I made one but it's fugly ... so I can't show it in public. I must admit it's the first model I ever unwrapped and textured properly ... but it's still fugly. What 3D app has a good renderer built in? |
| ||
| My try (746 polys)   dave |
| ||
| from my vantage point i find it interesting to see how forms migrate from person to person; how the idea of "crate" is so rigidly fixed in most everyone's mind. i'll also give credit to halo for the framework concept. just wondering who'll take it to the next level (kinda where i left off). throw out your ideas of what you know. Picasso once said it took a lifetime to finally get back to what children do instinctively. perhaps approach this from a child's perspective: not what would company XYZ fabricate, but what can you imagine. was going to model something, but this is far more interesting to watch and much more informative for me as an observor. thanks for that. |
| ||
Here's my entry, 12 Triangles, created in 3 Milliseconds, including texture: The texture creation is randomized using timestamp seed. |
| ||
Mine is child like ;) 109 verts, and 200 polys!    |
| ||
| Some interesting forms. I don't think you can ever think of anything "original". All you can do is recombine what already exists. |
| ||
| if i were a child, i might dream my wooden blocks could march around the room. arrange themselves to form words. stack themselves even (they're wood, let them grow twig arms). if they fell over, they could right themselves. if they tried real hard, they could even change the letters on their surface. some might even use they're splinters as torches to light their way on midnight forays to the kitchen! maybe not "original" in the grand defensive sense people fall back on, but then why limit this to what amounts to metal cubes? it's a container. a vessel. so is a balloon. replace regular rubber with the new MetalRubber and you have something amazing. double-wall it and insert a reactive liquid between the walls and you have a shapechanger which can be soft or hard. embed electronics in it and it could have rudimentary intelligence (which could even find it's way home like a dog lost on vacation). stop thinking like programmers and 3d application users. the problem isn't how to model a box... it's how to contain something and get it from point A to point B. |
| ||
| (stop thinking like programmers and 3d application users. the problem isn't how to model a box... it's how to contain something and get it from point A to point B.) I like the way your thinking here. dave |
| ||
| hahaha. you're right. and i could have said "engineers" too (from personal experience - college really put the blinders on me), but one would expect engineers as well as programmers to get to the basic issue like this, but so many don't. solutions seem to follow each other like lemmings. but the "3d application users" is still valid. too busy trying to master the tool to stop and ask what it really is that needs to be made. |
| ||
| Hey, I put electronics in mine :( See the keyboard and screen ;) You can type instructions to the crate and an on board system link relays information to the crane to tell it where to be placed! |
| ||
| i noticed. why type instructions? and why on the crate? many warehouse workers have either handheld or strap-on (that sounds so bad) data input/scanning devices. most electronics is not visually obvious and becoming less so. however, an emergency panel with a big simplistic interface might be appropriate. would simplify the texture too. you also put the keyboard on an angle (assuming the need for ergonomics, i guess). but if it's emergency use, it could be more integrated into the form. that could let you simplify this and focus on other things. the protrusions strike me as strange. the "spike" on top seems to rule out stackability. doesn't seem to recess into a snug groove. if i had to pay for warehousing, that would tick me off. the big loop on the side eats space when the crates are placed side-by-side... unless those fold down and integrate somehow. both implementations seem to be for crane/forklift use, but couldn't the crane have the complex geometry to grasp subtle crate details? and i esp noticed what i did take to be a scrolling billboard on the side. from Hazard warnings to selling the company shipping the stuff, there's plenty of logic in that (btw, catch the latest: "wheel advertising" - http://motortrend.com/features/news/112_news040819_adfleet ). i'd expand on that idea. imagine a disorganized warehouse with a hundred streaming banners of various types. wicked. |
| ||
| Hehe, I'll have to take a second pass at it. |
| ||
| a good reference to consider is those rolling military robots from Star Wars (Episode 1, i guess). the way they store in the "troop" carrier, how the storage rack interfaces to them and how they first deploy and then unfold is all nicely done. |
| ||
Here's one I just did in Wings: Not bad for me! 496 polys. |
| ||
| Simon: Now how many triangles is it? 'Cause I honestly think you're counting the quads and not the tris. |
| ||
| Simon, I would make the top ridges trianglular instead of square...hope that makes sense, I mean a cross-section of the ridge would be a triangle. |
| ||
| thats nice simonh |
| ||
| I like simonh's, it looks different yet simple and very practical. halo - your winch attachment looks far too thin and flimsy to be strong enough to carry such a heavy box. |
| ||
| It's made of an indestructable polymer. :P |
| ||
| We are proud to present the new interlocking magnet crate featuring unitrack technology from the United Federation of Science! This new storage technology allows multiple crates to be connected and transported via interlocking magnet technology. This new crate also carries over our previous unitrack technology so no cargo goes unfound. The UFS is proud to present this technology to provide a better future today! Competitively Priced at $349.95 ... visit your local UFS Cargo dealer today!    |
| ||
| hehe.. How much for the shipping cost to Norway? |
| ||
| That will be an additional $3,956,456,801,895. Would you like paper or plastic? |
| ||
| Jeremy, that's an interesting LEGO like idea. This promises stable storing . I guess the top side is carved then. But please use less or smaller pics. Uhm but wait, when there are 3 sides that are plugable, you won't be able to take a single crate out of the warehouse. |
| ||
| retractable interlocks instead? |
| ||
| Yeah of course they're retractable! The UFS is the finest design and production company in the mesosphere, what else would you expect from them! |
| ||
| Allrighty... my contest piece is finally here and even before the deadline... I made the texture from scratch afterall because I didn't want to reveal any "company logos" from my game etc possible spoilers. I used VERY simple forms and shapes because I wanted to have a boxy silhuette so that I can do stacks of them and remove all "non-seen" walls - you can't stack Halo's crate for example (like 3*3*3) and remove backfaces from it like you can from mine. Few "whys" "explained": 1. UV-mapping... I have tried to make the UV's very versatile, so that I can use the same texture for making loads of different looking crates, and by using vertex color tinting, few dozen more :) 2. There is empty space in the UV-map on the right side and this is intentional. I will use this space for small add-on bits, like shipping slip stuff and for various markings etc to make the boxes even more "individual". It's always good to have some room in the map for the "unexpected" need to add something later on. 3. Original texture was 1K * 1K which would be 256 pixels / meter resolution for the texture, and 512 version is of course 128 pixels / meter rez which is what I use for all my level stuff - it doesn't hurt to do the texture using 2X scale, because you can always downscale, but you can't scale it up once you've made it. 4. Since it was made for a game I modeled the inside also - what good are crates that can't have stuff in them? :P I also modeled one version that don't have the inside bits and is fully sealed (solid) for crates that don't need to be opened or for making rows and rows of crates (think aircraft hangar full of them or something like that). [edit] 5. Oh, and the crate is 1.5m x 1.5m sized because I wanted to have a crate that can give you some cover "in the unlikely case of a shoot-out in a fps game"... and you can shoot back over the crate too. Textured views, top off and on, total of 600 triangles (solid version has only 492 triangles):    Wireframe views, top off and on:   No textures, , top off and on:   UV-map:  Texture (512*512):  |
| ||
| Flipping heck Mustang! Thats awesome - looks real! IPete2. |
| ||
| That is one impressive looking crate Mustang, if you hadn't posted the wireframe and uv map I would think it a photo :) |
| ||
| Nice one :D dave |
| ||
| Very nice :) How did you make the texture look so realistic? |
| ||
| Sweet! |
| ||
| How did you make the texture look so realistic? It's just my style and work methods... I've been doing "photorealistic" textures few years now and I use as a base for most of my work real photographs (I even have two free packs of those on my websute, but they are currently down beacuse of the bandwidth they took). I wouldn't do "handdrawn" textures even if someone wanted, I think that we now have enough good hardware to do realistic looking textures too. This crate is still missing a good specular/glossiness map which would make it look much better and can be done in Blitz too [spherical reflection map and texture map with alpha on top that]. I have literally gigabytes and gigabytes of base/source images I have been taking with my and our company digital cameras over the years, and there are also few good places on the net where you can get royalty free photosources. These like most of my own snapshots are nowhere near "ready" textures, but then you fire up your PhotoShop and make them textures by tweaking like correcting the levels, hues, contrast etc and removing warpings... and I also use often only bits from photos and combine them to other photos, and use layers and layer effects extensively to get the final outcome I want. The crate texture has 14 active layers that all contribute something to the final result - keeping stuff layered is good also for making different versions of the texture like one without texts. Ideally you'd have only one base color (plain even color) and you built everything on top of that like the rust and scratches, logos and other stuff that makes texture look "real". This way you only have to change the base color to something else and you'd have a different colored crate - but try to adjust a brown rusty metal to be a blue rusty metal; it's hard because the base color and rust are so close to each other, you have to make a good mask by hand at least for area color change, but if you had the base color on a different layer... well you get the idea! Two (old) articles I wrote about modeling and texturing for real-time use that may make my approach more clear: http://www.caffeineoverdose.net/cod/articles_article1.shtml http://www.caffeineoverdose.net/cod/articles_article2.shtml |
| ||
| yeah... real nice 'Stang. that's where i'm trying to get to with my texturing... ... so, in the meantime, i'll use one of halo's free textures with a lil bumpmapping in Blitz, to present my entry :)  not much... just a textured cube, 12 polys, 20 vertices, 8 distinct ones... your basic low poly box out of Milkshape. it's so dark because i just finished playing a few minutes of Doom3 :) and i'm playing around, trying to capture that D3 look in Blitz. --Mike |
| ||
| My crate will win. |
| ||
| Jesus, my texture looks incredible. The hazard stripe looks like it is coming out of the surface. I am a genius. |
| ||
| easy enough to fix... a lil paint here and there on the normalmap... and waaa laaaa... now your texture makes you really look like a genius halo... along with a lil help from the code examples of some of the guys here... despite my lack of expertise in the texture making department... and my inability to make detailed low poly models... with a lil bumping here and there... i can fake it... all while keeping the polys a lot lower than i usually do :) --Mike |
| ||
   458 polys |
| ||
| Mustang has some competition! |
| ||
| i was wondering...whos judging this? |
| ||
| Man, it's all in the texturing ... jeepers ;) |
| ||
| " i was wondering...whos judging this? " Probably Halo, which means if I enter I have zero chance of winning, but at least I will enjoy the accolades of my peers. Halo already tried to judge the contest once at which point I pointed out that he was judging it early. And he refused twice to answer who would be judging it when I asked. :-) Perosnally I think we should take a vote at the end regardless of whether Halo wants to or not. This should not be a "make a crate that Halo likes" contest. :-) |
| ||
| I vote for my crate. |
| ||
| So far Fredborg's is my favorite. I was gonna go for a hand drawn approach, but it looks like if I complete my entry tonight, I'll have to go for photorealistic to compete. Oh and modeling crates in Max is a major pain in the ass. Click on an edge to split it and nine times out of ten, a vertex appears at some random location around the model, having no connection to the mesh, and thwarting attempts to undo the operation, requiring manual deletion. |
| ||
| now i no why i dont do texturing ;) dave |
| ||
| i'm gonna vote for Fredborgs too... no, wait... ... Mustang's crate, yeah... that's the one... ... wait, wait... uhhh, oh, they're both really top notch, i can't decide :) --Mike |
| ||
| But... I give points to Mustang for his creative UV unwrapping, and the fact that his crate has an interior. |
| ||
| I think everyone who didn't enter should be judging. |
| ||
| (I think everyone who didn't enter should be judging.) Nice idea EDIT .Just show the pics dave |
| ||
| Whoever won the last contest is the final judge. I juedge the first one. Don't worry, I won't pick my own. |
| ||
| I vote for mustang, then fredborg. A special price for lowpoly and quick production for myself :) |
| ||
| yeah mustangs is the best I think |
| ||
| Hey Fred, really nice crate you got there! Love the logo especially, very pro. Very compact mapping too. |
| ||
| Thanks :) Yours is nice too. The only thing I dislike about it, is that the big horizontal ridge (maybe not the right word) in the texture isn't consistent with the geometry. Apart from that it's great! |
| ||
| The only thing I dislike about it, is that the big horizontal ridge (maybe not the right word) in the texture isn't consistent with the geometry. Apart from that it's great! Yup, I thought myself too that it might be bit too much without supporting geometry... and I'm beginning to think that I should have made the box/lid seam corrosion bit less in-your-face... Idea was that it looks like something has/is oozing from the crate but the inside is too neat for that... I'll make a new version of the texture and/or the geometry tonight and correct these issues. Thanks for the feedback BTW, I always like to have that because then I can make my stuff even better. |
| ||
| Home Sweet Home ~ a crate of unparallelled efficiency and usefulness ~ BEHOLD!  ASTOUNDING TEXTURE MAPPING! (4096x4096 pixels)  AMAZING WIREFRAMES! (53,120 polygons)  The piece has 53120 polygons and the texture is of course, 4096x4096. It's cunning use of polygons and hand crafted uv mapping guarantee this topical piece a place in the crate hall of fame. |
| ||
| cool blood, have you used it with tokamak yet? |
| ||
| Yes, I had 500,000 million of them fighting in a huge space battle. |
| ||
| I would vote for fredborg's with the one by Mustang very close, but overall nice work everyone. |
| ||
| *********************************************************** *********************************************************** RESULTS *********************************************************** *********************************************************** 1st place Fredborg The mesh was clean, with some interesting details, especially the buckles. The textures fit the geometry nicely, especially the line where the lid meets the sides. I would prefer a more metallic look, and thought the buckles should be textured differently than the sides, but it was very good. How about posting the model for us? :D 2nd place Mustang Mustang has the best texture, although again, it is dirtier and more plastic than I was expecting. Regardless of the style, the texture is definitely well-done. The mesh is average, without much detail. Better mesh/texture interaction would improve this model, but the texture quality really makes this one stand out. Honorable Mention Noel Cower This is my favorite mesh, but without textures it can't compete with the skinned entries. Interesting, hi-tech design that would look great with some textures and cube mapping. The rules are that Fredborg picks the next contested item, due Sunday night, September 5th. Fredborg is the final judge, and cannot pick himself, though he can still enter. This message must be updated and included when he announces the winner next week. |
| ||
| Can't argue with those results! So, Fredborg, what's next? |
| ||
| I win with the tramp in a box. |
| ||
| I have a sneaky feeling my crate was overlooked? I think if "halo" is honest, that he will admit that he didn't even consider my crate - therefore, the result is void - all entries should have been judged. Still, I don't want to worry anyone but currently my chair is in the lead by one vote in the 'Make a 3D chair' contest! set by "Fredborg". |
| ||
| Oh, I judged it the minute I saw it. ;) |
| ||
| See, I told you Halo wasn't going to give my crate any awards! |
| ||
| Personally, I liked Arbitrages CRATE. |
| ||
| nice texture work. couldn't help wondering: why would crates in the future rust so much? materials scientists have just created alumina-based glass. have to imagine regular rust will go away. minute i saw Mustang's i imagined a different kind of corrosion; maybe something that creates a different color or pattern of sorts - but you know it's corrosion. just a thought. |
| ||
| "materials scientists have just created alumina-based glass. have to imagine regular rust will go away." Heh. You don't need space age materials to get rid of rust. Rust is created when oxygen combines with iron. Steel has iron in it, so steel rusts. But aluminum does not have iron in it. Aluminum oxidizes and will get a white powdery coating on it, but it's not at all like steel rusting. :-) Of course Aluminum is not very strong, so you might not want to use it for crates. But then, why would you want to use any metal at all for a crate? Metal is heavy, and some kinds rust or oxidize, and it condustcs electricity. Doesn't seem a very good choice. So why not plastic? Well steel is stronger than plastic. So I suppose we couldn't use it for very large shipping crates. But I imagine that one day we will have a space age plastic that... Oops, went right back to the space age stuff didn't I. :-) I'm sure we have perfectly good materials for making crates that don't rust, like Titanium, and/or are light... like, perhaps, carbon fiber. But these things are expensive, and we tend to make stuff from the cheapest materials available. :-) |
| ||
| forgive my simplicity. if we want to discuss materials technology, i'm certainly up for it. perhaps we should start a new thread to discuss the oxidation process, the variety of secondary operations to eliminate it, or even get into the crystalline matrix issues that allow it to occur in the first place. or we could get into buckyball options that are opening up a whole new class of materials. or maybe.... wrt titanium, i guess people haven't watched the movie "Outland". guess what they were mining on Io in that SF flick.... wrt carbon fiber, that cost is prohibitive both in the material and in the lay-up. i assume everyone here is aware of why the lay-up issue is going away and composite materials are ready to make a leap forward in useability.... does it really require a treatise to get my point across? |
| ||
| Idea was that it looks like something has/is oozing from the crate It doesn't have to be the CRATE... have you ever seen batteries that spill out their content? :) |
| ||
| i read that earlier. but it's such a nice job showing corrosion.... on first view (before that post) i didn't think of something oozing out. maybe because you have the red on the lid above as well. i do think the oozing look would be cool tho. |
| ||
| Csven: Hehe. :-) Hey, if you know more about materials than I do, that's cool, I'm always willing to learn new stuff. :-) I just told you what I know. I don't know anythignabotu any crystalization stuff, or what issues there are with carbon fiber. I heva heard of buckyballs though. I've also heard that they could be dangerous if they get in your brain and cause degeneration so we need to be careful with some the weird new materials we're creating. :-) I also know that carbon tubes will explode if you flash a bright light at them. :-) |
| ||
| fair enough. the carbon fiber thing is cool tho. there's a Japanese fashion designer (yes, a fashion designer) that hired a guy to help him create a way to weave in 3D. most looms are obviously 2D. well, it works. and he's getting some patents for his 3D weaving process. and a ton of interest from all sorts of different industrial sectors. read about a week ago (some story about the remote air drones the USAF is developing) and some guy at the defense contractor mentioned new developments in the carbon fiber body-creation process. almost certain this is that fashion designer's process (think he mentioned military interest in an old interview). check out the links i posted on the "Make Chair" thread. ever since my aerospace days i've wondered when we could fabricate a wing like a bird bone. there's one chair shown that tells me it's coming. and may be already on the drawing boards. |
| ||
| I think I heard about that 3d weaving thing... I wonder how that works... I can see how weaving a tight 3d block of carbon fiber might be quite useful. There's something else interesting I heard a Japanese guy came up with regards to strong materials. Apparently he did something like put glass fibers in latex paint, and when painted onto a concrete surface it greatly increased the strength of the material. I think they were talking about other uses for it too. I thought it was odd that simply putting random threads of glass into latex paint could cause it to gain so much strength. As far as bird bones go... I don't know exaclty how those are structured, but if I'm not mistaken, they're porous right? And hollow? I always thought that hexagonal structures like a beehive would make good materials. I think I heard they use some stuf flike that in airplanes. I wonder if there is another shape which could tile in 3D... Would that be a dodecahedron? Perhaps there is no 3D um.... standard... shape beyond tetrahedrons and cubes which can tile in 3D, I'm not sure. |
| ||
| bird bone i saw looked more like if you glued two pieces of something together and then pulled them apart a bit. all stringy in between. very organic. closest thing i can imagine that we're doing like that today is gas-assist injection molding. but the cross-sections i've seen are just variable wall-section; not stringy like the bone. which is why that one chair structure caught my eye. i was recently sent some samples of some new SLA material. it felt a little like polypropylene.... and it flexed. SLA materials have traditionally been really brittle. i could conceivably create a wing in 3D and rapid-prototype the part just like that chair. now i think about it, i may have another option using a common manufacturing process. have to think about that some. |
| ||
| I thought halo's was the best. We were grading for best "futuristic" crate, weren't we? |
| ||
| My votes in order: Fredborg Mustang Halo |
   |