Sensei Challenge No. 5 - Greebles!

[Jedilaw]

The SFM Sensei are proud to announce the latest Sensei Challenge (Challenge No. 5): Greebles!

As most (if not all) SFM members know, a "greeble" is a detail piece added to a larger model. Specifically, greebles are cool looking details with no apparent function, things that look good but don't actually do anything. Greebles can add a suggestion of apparent depth to a model, and they can enhance the sense of scale, but mostly they just look good. Greebles are there becuase they are interesting to the eye. Think of the Millenium Falcon: it's covered in cables, pipes and equipment, and nobody knows what any of it does. In a word: greebles.

The object of this competition is to see who can produce the most impressive greeble. The greeble can be composed of virually any shape, be it boxes, tubes, pipes, or spheres. Because greebles are a wide-open category, the judging for this competition will necessarily be somewhat subjective. The Sensei will be considering the following criteria, among others:

1) Level of detail achieved (how far beyond a chamfered box have you taken your greeble?)

2) Originality

3) The "cool" factor. By that we mean the degree to which the greeble makes you immediately want to say "man, that looks really cool!" In other words, is the design something that is immediately striking?

For inspiration, you might consider downloading one of the greeble sets available in SFM's mesh section (keeping in mind that any submission that is simply taken from an existing set is disqualified). You may find inspiration in images of various well-greebled models such as the various Star Wars models, the "Roger Young" from Starship Troopers, et cetera. The following sites have excellent reference images:

CultTVman's SF Modeling Page (http://www.culttvman.com/) -lots of images of studio models, or close replicas of them
Starship Modeler - Your Complete Information Source for Science Fiction, Fantasy and Real Space Scale Modeling (http://www.starshipmodeler.com/) - ditto
Forbidden Plastic (http://www.forbiddenplastic.com/) - for hardcore scale model folks. Good for reference
Cloudster.com (http://www.cloudster.com/EntryCardSets.htm) - excellent shots of the TMP Enterprise model, the TMP K'Tinga Class Klingon cruiser, and the 8' Star Destroyer model used in the Empire Strikes Back and Return of the Jedi.
Studioscale.com (http://www.studioscale.com/kitscans.htm) - images of Tamiya, Airfix, and other model kits - useful when modeling greebles
Meshweaver (http://www.meshweaver.com). This is Fabio Passaro's site. He modelled the engine section of the new Galactica, and has put together a greeble (or "nurnie") set for commercial sale. You can view an HTML catalog of his nurnies, several of which should provide inspiration.

Here are a few rules for this challenge:

1) The greeble must be entirely your own work. In other words, you cannot simply modify a piece from an existing greeble set, nor can you combine multiple pieces from an existing set.

2) No greeble plugins or scripts are allowed, neither is any premade geometry (basic primitives are fine to start with, of course).

3) Your final image should be an un-textured "clay render." We want to be able to judge the modelling on its own merits, not affected by textures or lighting effects. (EDIT: the requirement of only one light source has been removed, at the request of our members. Hopefully, the rules are more workable now).

4) Final render must be 800 x 600.

This challenge will commence on Monday, February 12. Deadline for submissions is midnite on Monday, February 19. By midnite, we are referring to midnite in your local time zone, Monday night (not Sunday night).

[Melak]

So are we supposed to model "one greeble" as in one little piece( like a single part from one of your greeble packs), or are we allowed to detail a surface with as many greebles as we want?

[Jedilaw]

Make one greeble. It can be complex, as in a combination of many shapes, but it needs to be one object.

[DarthMaya]

aww I suck a greebles, o well I will try. :D

[Jedilaw]

aww I suck a greebles, o well I will try. :D

There is no try. Do or do not.

j/k

[DarthMaya]

There is no try. Do or do not.

j/k

True..... true.......:thumb:

[Tinukedaya]

This might be an interesting go. Although I suck at greebles. <grins>

One thought for the rules though... Wouldn't it be nice to somehow set the "size" of the greeble? I know it'll be hard to judge anyway, but something like Make a greeble that fits 2x1 meter hole or something like that.. It might uniform the scales a bit and help the judging.

Or well. scratch that.. someone might still put in some micro-industrial-complex and that will break the scale as well. so do not mind me.


On more serious question.. I will be offline at that week, although I'll still be able to do some modelling.. Now the question is... I will not be able to post my progress, Can I still post the final submission? That is if I'll do something useful and manage to get online before the deadline?

[aszazeroth]

Can we use displacement maps for part of the geometry !? or does it have to be "pure" geometry based ?!

[Jedilaw]

This might be an interesting go. Although I suck at greebles. <grins>

One thought for the rules though... Wouldn't it be nice to somehow set the "size" of the greeble? I know it'll be hard to judge anyway, but something like Make a greeble that fits 2x1 meter hole or something like that.. It might uniform the scales a bit and help the judging.

Or well. scratch that.. someone might still put in some micro-industrial-complex and that will break the scale as well. so do not mind me.


On more serious question.. I will be offline at that week, although I'll still be able to do some modelling.. Now the question is... I will not be able to post my progress, Can I still post the final submission? That is if I'll do something useful and manage to get online before the deadline?

I see no problem with you posting a final submission without having done a WIP thread. WIP is not a requirement.


Can we use displacement maps for part of the geometry !? or does it have to be "pure" geometry based ?!

Please refer to item 3 of the rules: Your final image should be an un-textured "clay render" with a single light source. We want to be able to judge the modelling on its own merits, not affected by textures or lighting effects.

[aszazeroth]

Sorry about the confusion... a displacement map can be clay-rendered too. Guess I can always "bake" in any of the displaced details too, so the question was perhaps unnecessary.

[gonk droid]

great idea. i'm in.

[Pyrocitor]

Great idea

I might have to give this one a shot, I've already got ideas flying around in my head :D

[Ozylot]

... Your final image should be an un-textured "clay render" with a single light source. We want to be able to judge the modelling on its own merits, not affected by textures or lighting effects.

^DOH!

I would have loved to take this one on.

But, I cant do bounced light. and 1 light just wont cut it.

[Jedilaw]

^DOH!

I would have loved to take this one on.

But, I cant do bounced light. and 1 light just wont cut it.

Hmmm, if it's not technically possible for you to conform to the lighting requirement, an exception may have to be made. I'll discuss with the other Sensei.

What app are you using?

[Coolhand]

Ozy, just use whatever lighting you need to get a nice clay render, jedi, can you remove the bit about 1 light source.

[Ozylot]

I use wings3D... and I can get decent renders out of it using yafray, but its much more difficult. Thus the reason I use bryce.

I use bryce for rendering, and in order to get 'simulated radiosity' I need to use a ring of lights. (the more the better quality. 72 at most). If that restriction is negotiable, I'd love to enter! Thanx for the understanding.

[Forcemaster2000]

Sounds fun! I may just be able to get into this one!

[JustinDixon]

ohh, fun. I'm in.

[Tovette]

It sounds like a fun challenge. Although I am curious- is there a polygonal cap to how high someone can build their greeble up? I think practicality of polygon use should be considered. I mean if you build a greeble that's 500,000polys then what use is that? Know what I mean?

[Sphynx]

We did not set a cap as such, but remember that as part of the judging we are looking at the skill you have at building the geometry.

As a greeble is traditionally meant for multiple use, personally, I'd say don't breach 20,000 faces (thats NOT an official limit), but also remember that you need to use them well. If it takes you 8,000 polygons to make a hemi-sphere in the middle of greeble, then that's probably not a great statement to make to the judges.

[Tinukedaya]

We did not set a cap as such, but remember that as part of the judging we are looking at the skill you have at building the geometry.

As a greeble is traditionally meant for multiple use, personally, I'd say don't breach 20,000 faces (thats NOT an official limit), but also remember that you need to use them well. If it takes you 8,000 polygons to make a hemi-sphere in the middle of greeble, then that's probably not a great statement to make to the judges.


Might be wise to ask for wireframes as part of final submition, for you to see the geometry maybe, uh?

[Pyrocitor]

How big and complex can the greeble be?
What I mean by that is can it be a similar compexity to the two left greebles (in the image) or more like the two right greebles - and yes those are Jedilaws greebles

If we did have a Poly limit then you wouldn't have to wory about the complexity as you would have to make a trade off between quality and complexity.
And that would to a degree help you judge the contest, as someone who's really efficient at modelling would be able to make a super complex quality greeble.

Ok, that was alot shorter in my head :D
Oh yea, and I'm guessing it's this rendering style your refering to

[Enterprise-E]

i think this would be more interesting if we would be given a basic hull shape that we must cover in greebles. Sorta like pimp my ship, just this would be "detail my ship to hell". Making just one greeble is meh... and will end up with greebles actually unusable because the huge polycounts. Every one will try to add more details than the others. a single greeble, irrelevant how complex would never be even close to 20k polies, because it would kill every model where it is repeatedly used.
Also, judging a single greeble is VERY subjective IMO, while judging the end effect of what we made from a predefined hull shape would be something more objective.

just my 2c

[Sphynx]

Nope - the contest is running (in theory) with people thinking about what they are doing ready for offical start on Monday. Its really up to you guys to get the best value for your polys - impressing us does not just mean giving us a great greeble remember, but also making it class as a greeble - as you've said E, too many polys and it becomes useless as a 'greeble'.

Impress us.

[Jedilaw]

i think this would be more interesting if we would be given a basic hull shape that we must cover in greebles. Sorta like pimp my ship, just this would be "detail my ship to hell". Making just one greeble is meh... and will end up with greebles actually unusable because the huge polycounts. Every one will try to add more details than the others. a single greeble, irrelevant how complex would never be even close to 20k polies, because it would kill every model where it is repeatedly used.
Also, judging a single greeble is VERY subjective IMO, while judging the end effect of what we made from a predefined hull shape would be something more objective.

just my 2c

That same idea has occurred to me, but I am keeping it in mind as a possible later challenge. It might consist of something like "here's a basic box shape, here's a set of greebles, show us what you can do." There are other related challenges under consideration, which we think the members will enjoy.

As for poly counts and repeatability, not every greeble shows up dozens or hundreds of times. Some have only a handful of instances. Think of the more complex greebles on the Millenium Falcon: the really involved ones show up a lot less frequently than the more basic ones. Or think of the dome-shaped apparatus on the bottom of a Star Destroyer: it's definitely a greeble, as it has no known or apparent function (some say tractor beam, others say different) but it looks damned cool. And modeling it properly would be massively poly-intense, I would expect in the range of 100k or more. It wouldn't necessarily nuke a project, though, because it only shows up once on the ship.


How big and complex can the greeble be?
What I mean by that is can it be a similar compexity to the two left greebles (in the image) or more like the two right greebles - and yes those are Jedilaws greebles

If we did have a Poly limit then you wouldn't have to wory about the complexity as you would have to make a trade off between quality and complexity.
And that would to a degree help you judge the contest, as someone who's really efficient at modelling would be able to make a super complex quality greeble.

Ok, that was alot shorter in my head
Oh yea, and I'm guessing it's this rendering style your refering to

Yes, that's the "clay render" style we are referring to.

As for complexity, go wild. Speaking as the creator of those four greebles, I take more pride in the ones on the left than the ones on the right. In fact, the ones on the left are among what I consider to be a handful of my best publicly-released greebles. That is only my opinion of course, and it's not meant to suggest that a more streamlined greeble couldn't win if it looked striking enough.

You may have noticed that the more complex greebles are in fact combinations of other greebles that were also included in the released sets. I tend to create high-poly greebles by mixing and matching components until I have something that looks complex and interesting to the eye.

BTW, if you're looking for more inspiration, try entering terms like "porsche transmission" "Ferrari transmission" "automobile parts" or "engine parts" into Google Images. Some interesting stuff comes up.

[gonk droid]

wondering if sunken greebles are allowed? where some details start below and work up/out.

http://i156.photobucket.com/albums/t31/gnkdroid/inset.jpg

[TheMightySpud]

I like this idea, I'm not gonna win by a long shot, but I'm gonna have a go, hopefully learn something in the process :-D

TheMightySpud

[Jedilaw]

wondering if sunken greebles are allowed? where some details start below and work up/out.

http://i156.photobucket.com/albums/t31/gnkdroid/inset.jpg

Sure. The structural criteria are wide open.

[gonk droid]

Sure. The structural criteria are wide open.

thanks.

[farcry]

Im in :D

[Kadaeux]

I'm most certainly in! :D

[toby-wan]

so this is a go now? sounds like fun.

[Jedilaw]

Yup.

[farcry]

OMG it has started!! *does face like Jedi's avatar* :D

[Ozylot]

Pardon me if this sounds silly, but where does one post a wip for this?

[Sphynx]

Post in the root of the Challenge areas - currently, you'll just see a message saying 'no threads in this forum' - just create a new thread.

[Sphynx]

In expectation of the end of this challenge, and based upon PM's that have been made to moderators, please note:

3) Your final image should be an un-textured "clay render." We want to be able to judge the modelling on its own merits, not affected by textures or lighting effects.

4) Final render must be 800 x 600.

A wireframe render is NOT required as long as we can see your progress in a WIP thread.

When you are ready, post your images in this thread (http://www.scifi-meshes.com/forums/ring/17286-sensei-challenge-5-post-your-final.html).

IMPORTANT: If you have questions about any challenge, please post in this thread of PM a SENSEI - please do not PM Moderators as they can not answer your questions and will just need to get in touch with one of us. To speed up the process, just come to us FIRST.

[hal]

I'v I have no work today and have a free day of no client work so I'm going to give this a shot! question, can you do a greeble of a city block?

[Sphynx]

As no sensei had been on a for a while, Hal PM'd me - to be fair to all members entering the challenge (despite the fact, that we're in the last day ;) ), the answer that I gave him was:


...From my point of view, I don't see why not as it conforms to the standards that we set - whether the repeating unit is attached to a starship or a city, to me is irrelevant.

As long as it satisfies all the ideas of a 'greeble' - ie. it's not, by definition, unique and could clearly be re-used within the same set (e.g. not many cities have a dozen, repeating Royal Albert Halls or Empire State Buildings for example)....

[gonk droid]

edit: my mistake. hadn't set my time zone preference yet.