First time poster
polycounter lvl 13
Alright guys, my first post here. Figured i should get my early wip up here before The Keen does. Anyways, project off a concept sketch (artist unknown - from concrete games). 5000 tri limit.
3ds maxed so far, still need to model out details and what not, only 2,600 tris into it. . .should be textured in 2 weeks.
oh and does anyone else think this looks weird, the v shaped crack right in the middle?? just wondering.
and the back side - which i'm still trying to figure out if anyone has any ideas shout somethin out.
3ds maxed so far, still need to model out details and what not, only 2,600 tris into it. . .should be textured in 2 weeks.
oh and does anyone else think this looks weird, the v shaped crack right in the middle?? just wondering.
and the back side - which i'm still trying to figure out if anyone has any ideas shout somethin out.
Replies
I can tell already you are an AI-San Diego student. Please for the love of god wipe this off the transfer drive at school.
First of all it is Concretes ART TEST and if they didn't send it to you I don't think it is right that you have it or anyone else for that matter.
Second you are an artist and doing this just shows that you are not original at all.
Third I take it you are going for environments / props / etc. This piece is really limiting on the specs. While you can get good results you are putting yourself up against everyone that has already done this piece and shown it.
Ask yourself. Am I gonna be able to make it better than anyone else that has done this?
It's just my opinion but honestly it's not a good portfolio piece and that is what you should be working on.
The V looks weird but only because that one plywood shape repeats so often. Plywood is a fix, it doesn't stop bullets well and if you look at the ref more closely you'll see that is meant to be the remnants of the original railing, not shanty board cover. This is a chance to show hints at what the building was originally build for, show some history in your prop and help sell the hap-hazardness. If you're set on using plywood as a railing, you should bake that one corrugated metal piece into a normal map so you can use it to better break up the repeating plywood without having to worry about it costing more polys. You can even hide the ends under the other boards so the obvious normal mapped ends won't be obvious.
I think you need to work on the destroyed corner of the building, the floating chunk looks too big and odd.
In the ref the window on the corner is blown out and it is a great opertunity for detail, but I think you're wasting it by just chipping off a thick layer of plaster/brick. You should open that corner up and show some of the inside, maybe add sand bags like its a gun encampment. "OH GOD I HAVE TO MODEL THE INSIDE!?" No just the parts you'll see from the outside it can be done very simply but can make a huge impact on the silhouette and level of detail.
Next put yourself in this scene, live here, work here. Come up with some back story for your dirt, debris, and clutter. Think about what you would need, use and have if this was your place. How would you keep it safe when you're not here, how would you defend it when its under attack? How has it been attacked before and you have since had to repair or improvise?
Lastly give your props personality, instead of making "barrel991a" and placing the same prop in a cluster 5 times. Make "HappyBarrelDarrel1" and "SadBarrelSheril1" get that emotion across and then build a stack of 5 barrels with them. 1) it will be more interesting to look at and different from everyone else. 2) it will give you different details to work with as you rotate the prop giving the player the impression you have less repeating props than you actually do.
It looks like you're off to a good start, I like what you have even if the subject matter is a touch over done. It can be a great piece if you push it.
eeeh, well my reputation is already ruined. .. .
[/ QUOTE ]
No it's not. Everyone here is only trying to help. Honestly if you can make this piece look amazingly cool, go for it. Your ability to model solidly from a concept and paint interesting and smart textures is what will get you a job, no matter what the subject matter.
eeeh, well my reputation is already ruined. .. .
[/ QUOTE ]
>>Snap goes the rubber glove<<
Polycount 101:
Well thats what's different about polycount, most of us don't coddle people. Pimpin&Previews should really be called "rip my art apart and tell me how to make it beter" but the name is too long heh. Most of the "harsh" crits you'll get here are aimed at helping you take your art to a new level. They may be hard to read but they are constructive. They are exactly what you would hear from any Art Director. If it wounds your ego to read such things think about how its going to feel when those same comments come up in person, in a room full of other people, during a in a build review. You'll either need to grow some thicker skin or consider a job where people don't ask you to push yourself.
Things to think about:
Some places are fine with people who just tow the line and make the shanty they see in the ref pixel for pixel. But honestly those places aren't going to be pushing any boundaries they're just happy someone applied.More than likely you'll be working on Barbies Horsie Round-Up for 5 bucks below min wage, in North Korea, where the strength of your power supply is based on how fast you can pedal.
If you're happy working at a crap hole place, and you're happy just towing the line then I'm happy you're happy, ignore the suggestions we make.
Based on what I see I wouldn't suggest making improvements if I didn't see something I liked. What you have is "good" and would work in a few games just fine, but I think you have it in you to do better if you push yourself and consider a few things in new ways. You can try to get a job with this shanty the way it is and maybe have a few doors shut on you but eventually one will open, or you can push it and have that door open sooner and have it lead to a nicer place to work. Really its up to you and how much you're willing to put in that determines how much you get out of the place you'll be working at.
Do you want to beg for an interview, then beg for a job, or do you want them to beg YOU to work there? I know what position I would like to be in...
[ QUOTE ]
eeeh, well my reputation is already ruined. .. .
[/ QUOTE ]
>>Snap goes the rubber glove<<
Polycount 101:
Well thats what's different about polycount, most of us don't coddle people. Pimpin&Previews should really be called "rip my art apart and tell me how to make it beter" but the name is too long heh. Most of the "harsh" crits you'll get here are aimed at helping you take your art to a new level. They may be hard to read but they are constructive. They are exactly what you would hear from any Art Director. If it wounds your ego to read such things think about how its going to feel when those same comments come up in person, in a room full of other people, during a in a build review. You'll either need to grow some thicker skin or consider a job where people don't ask you to push yourself.
Things to think about:
Some places are fine with people who just tow the line and make the shanty they see in the ref pixel for pixel. But honestly those places aren't going to be pushing any boundaries they're just happy someone applied.More than likely you'll be working on Barbies Horsie Round-Up for 5 bucks below min wage, in North Korea, where the strength of your power supply is based on how fast you can pedal.
If you're happy working at a crap hole place, and you're happy just towing the line then I'm happy you're happy, ignore the suggestions we make.
Based on what I see I wouldn't suggest making improvements if I didn't see something I liked. What you have is "good" and would work in a few games just fine, but I think you have it in you to do better if you push yourself and consider a few things in new ways. You can try to get a job with this shanty the way it is and maybe have a few doors shut on you but eventually one will open, or you can push it and have that door open sooner and have it lead to a nicer place to work. Really its up to you and how much you're willing to put in that determines how much you get out of the place you'll be working at.
Do you want to beg for an interview, then beg for a job, or do you want them to beg YOU to work there? I know what position I would like to be in...
[/ QUOTE ]
how about ASS Rape 101 ?
I still say there needs to be more people who rip on here with helpful advice though, seems like a handful of people who give full and good critique other than, it needs work.
But yeah man as far as your scene goes I suggests maybe a bit more mounds of garbage and such around, make that thing have a few misc pieces around to add to the effect of beaten up building, newspapers and soda cans all that kinda stuff
There's 2 sides to everything:
It's fine to base your work off other people's concepts. Actually, unless you're aiming for a lead position it's a terribly good thing to show that you can work off other people's concepts.
Beyond that, it's just as important to how that you're able to improvise and expand a concept, photo ref or desired style, ie: take the style and feel given and build a cathedral or town plaza to match it if necessary.
So, in a portfolio you want to display both skills, the ability to read and interpret a concept or idea properly and the ability to match, improvise and expand a desired style and look.
As for pushing yourself and doing your own thing blabla... until you master anatomy perfectly, you'll only make a fool of yourself if you try to be 'original' and don't base your work on reference. Meaning that, until you master the trade, you should learn to mimic what you see and learn why it heavens it works and looks good.
I don't care how original something is, if it looks OFF.
As far as your building goes, i think it's coming along pretty nicely but you kinda got the railing stuff wrong i think. Yours looks all broken and replaced while in the concept, 1 side seems to be undamaged (if cheaply constructed) while the other side broke down and has been completely replaced with a wooden frame and rusted metal planks.
You can either say you got it wrong, or are trying to be original
Spark
Don't take things to heart. We will sometimes say things that seem like your work is garbage and you should apply at your local Burger King but we all mean well.
This place will seriously push your work to a higher level if you listen to what people are saying and don't get offended.
I also just found out that this is now a class project at AI so if thats the case then yeah you should finish this. Just make it totally bad ass. Add stuff to the environment around it. Trash cans, fences, busted up crates.. Anything that goes with the scene.
Good luck and looking forward to your work in the future.
Some real gems out there. Sucks for those of us who actually did it within a deadline and did a decent job. Now every Tom and Dick out of AI has this thing. Now I gotta make something to replace it
- yes this is a class project so i have to do it, i'm just sharing my newbness with you all in hopes of recieving good ideas and direction.
- yes i'm another ai guy. . .
- but i dig poly count and anything you all have to say is appreciated.
Nah, I'm just messin, do what you want.
Fit all 2 stories top to bottom..the ground and sidewalk..all the little pieces into a 1024 texture. Do a great job on and then tell me it was easy. Normal mapping something with those sort of specs is also a royal pain, because you can only overlay certain things a certain way...
Either way, I would suggest doing it. If you don't get a decent product out of it, it might be an eye opener for where your portfolio stands.
Getting into Emer's class and seeing 95% of the class couldn't get a single prop modeled and unwrapped in 4 hours, let alone textured was just mind boggling to me. Makes me wonder what some people think they are going to do when there is something called a deadline. Oh wait, they won't be in that situation. Their portfolio is full of clay renders and unfinished mudbox globs because that's "next-gen".
I miss school.
I love doing ruins and stuff like that
It's a boring, very basic art test. You wont learn much from it unless you completely rock it. Add some characters hanging off the roof, or trying to install the dish on top, or put a crashed helicopter in the side of the thing. Get the creative juices flowing, go above and beyond the boring concept.
OR do what you want and to hell with us is the typical response.
I do have to tell you that your $80,000 education (this is not an exaggerated number) ripping off a companies art test for class material is pretty fucking lame. I go to Ivytech in IN, and we have themed exercises but never rip a test from a company. WTH are you paying for anyway. You can just grab art tests, and model them from home no?
Thankfully there are a few teachers who go above and beyond, but sadly only about 5% of the faculty has any real game experience within the past 5 years. The administration side is even worse. Most of them don't know the first thing about what the degree is even about, let alone the needs of the industry.
And if you were going to OC it would probably be closer to 90g.
The only thing I did get from school was a lot of connections which in this industry is key but a lot of those same connections were made at conferences or through cgtalk, polycount.
The schools classes really are not challenging to say the least BUT you get out what you put in and some people that go there feel that they will get a job just because they went to A.I.
Haha....
-Wireframe. It's hard to crit this without seeing how it is put together.
-Sat Dish... too deep. look at some reference on this.
-Damage to the building doesn't seem realistic. If it's a brick building the edges shouldn't be as jaggy as not all the bricks will break like that. Some will keep the brick shape and you should try and use that as well.
-The piece of the wall that is just floating. Right now I'm looking at this and thinking how is that still there. If you look at how thick the walls are then this would have walls like 2-3 feet thick. That is way too much.
-hanging wires. how many tris did you spend on these? I don't think that is where you should focuse a lot of tris for this to make it interesting.
-Scale. Add a biped in that is in scale to your door and see how he looks by moving him around in the scene and compare him next to some things.
right now some of the scale seems off. The overhang over the garage doors looks to be really low. Like maybe 6 feet off the ground.
Keep in mind that when this test was given to the people that were interested in Concrete they had about a week to complete it so while you shouldn't rush your way through it and miss important things don't spend a month working on it either like some people do.
So far so good. Just keep up the progress and post some wires.
--also, i'm already workin on uv's . is it possible to rework my mesh without having to redue all my uv unwrapping?
Welcome to polycount, where what you put in is usually equal to what you get out of it. Meaning, that if you listen and work with critiques given, and work hard then you will be better for it. There are to many people that take critiques as a personal attack, which you need to realize that as an artist, critiques are not an attack but either advice or an observation. I agree that this piece is going to be in alot of portfolios, and you can either decide to push beyond what the concept is showing and make it more of your own, like Vig suggested. Or start on something new, like Jesse was in a not so subtle way suggesting (
Spark
[/ QUOTE ]
Am I gonna get a suit now , mens wear house?
haha yea I agree listen to this guy, he knows what hes talking about!
JesseMood: Yea AI im going to is in the same boat, I say theres about 3/4 who post on here in the entire school that I know of, and other than that everyone else seems to be fine thinking they will get jobs with the schools help. there writing off my buddy who got hired at midway as a success story the school took credit for, what there leaving out, is his art tests, the time he took off work the last 2 months, him going to GDC, and wasting his own money he had been saving just to meet contacts....gotta love school edumactionz
your scene seems to have ALOT of tris for a flat plane, whats up with that, you boolean something there?
ground as well, did you model that chunk of garbage from the floor itself instead of its own unit?
tighten up the graphics on level 2
dont use the ways of the boolean for everything
welcome to polycount! and do enjoy getting ass raped, and having 'new ones' ripped for u! cuz a) u'll never have constipation problems in the future, and b) these guys here give genuine crits. i swear i'd be nowhere now if it wasn't for these guys. wait... i am nowhere.... but hell, i've improved faster in 1yr wit these guys than i did in 4years with some other forums, and alone. they won't bullshit u about how nice ur art looks if there's something wrong. they're honest, and that's what helps most. even though those discussions ended a few posts above, i thought i'd jump in on that wagon too
btw: yeah, it seems like u went crazy with some booleaning? that always screws meshes up i think. so u'd have to go through a cleanup before uv templating or something along those lines.
cheers!
Just looking where your stairs meet the wall i can see a ton of wasted polys and where the building meets the ground.
The grounds flat. Why waste tris to get the shape of the ground when all it is is flat?
You can use interpenetrating geometry and it would save you so much time when it comes to optimizing and uvs.
I'm guessing you are wasting a couple hundred tris or maybe even more.
http://boards.polycount.net/showflat.php?Cat=0&Number=226128&an=0&page=3#Post226128
You need to start cutting up the geometry so that you can overlay the UVs. The same will have to go for a large number of the objects. Sometimes you have to make intentional cuts down the center of objects so that you can overlay the UV's in order to maximize the texture resolution.
Oh, and Jesse, you are going to love this - but some of the geniuses at AI (Ya know, the ones with zero industry experience) will get a single industry person to tell them that they use closed geometry in their engine, and therefore the teachers demand it in their class...never teaching about floating geometry, and basically telling students it is bad and "not used in the industry."
Yeah Ott is right about the ground textures and the other pieces. For this you really have to be amazing with your uv layout and not waste anything. Overlaying is key and so is mirroring anything you possibly can.
Yeah I know how it is at AI my friend. Sad...it really is but hey look at the bright side we (Gavin, Phil, Geoff, Judd, You, Me, Adrian, Tyler, Artur, Nate, Carlos and I know I'm forgetting a ton of people) all have pretty solid portfolios and good jobs.
Then again we never really did listen in class and kinda did our own thing but yeah. Floating Geometry not used... Shit I better tell pretty much every company I have done contracts for lately that I have a shit ton of models to fix.
quick question - is overlapping uvs ok w/ normal maps? i've heard its a big no no.
as for floating geometry, i have yet to hear that phrase get muttered in school and yes its been a lot of "water tight" labor and headaches.
You won't possibly be able to lay out that ground area and still get a decent texture quality with a 1024. That is probably the hardest part of the exercise - still getting the ground to look halfway decent while fitting the entire texture onto a 1024.
[/ QUOTE ]
@ ott,
Last friday, by accident I found nice collection of your gallery full of awsome props, and I saw the same project you have finished. So I guess you know this project thoroughlyk, don't ya?
I'm wondering, if you can give a bit more detail explaination about how to manage overlapping UV in order to maintain decent ground texture. How you deal with normal mapping then? Do you move the overlapping part off out of 0 to 1 UV space and bring them back, afterwards? (well, this is I learned from ben mathis' tutorial...)
I am off school now but still learning max by myself in searching for a job..
thanks in advance.
@ breakneck,
I ran across this tutorial the other days, and guess you may be interested in reading it.
Especially 'how to deal with overlapping UV' in baking Normal. Well, I could be wrong, since I am still learning Max by myself. Hope this help though.
[http://www.poopinmymouth.com/tutorial/normal_workflow_2.htm]
Rybeck
is overlapping uvs ok w/ normal maps? i've heard its a big no no.
[/ QUOTE ] It really depends on the engine you're working with. Most engines it doesn't matter, and it will only get screwed up when you bake. BUT there are ways around that. Such as moving one of the pieces up in the W space, or moving it over like Ben Mathis covers in his tutorial.
I think you need to look at optimizing what you have using less joined geometry. Then look at cutting your geometry to make use of tiles and save some texture space for big detail items. Everyone concentrates on poly count but texture memory is often a bigger concern.
[ QUOTE ]
as for floating geometry, i have yet to hear that phrase get muttered in school and yes its been a lot of "water tight" labor and headaches.
[/ QUOTE ] Kill these people, honestly they are ruining so many people it's scary. They need to be stopped. Every other artist I work with came from an AI and I've had to teach them new ways to work. You know how many times I heard "but at AI they said" or "yeah but in Maya its called".
some instructors... they just don't know is all. they mean well.
As for crits yeah gotta echo what Vig and everyone else has already mentioned.
You are gonna want to be able to overlay this stuff as easily as possible so even cuts here and there is what you really need to do.
Also don't worry about the overlapping. You aren't going to be building a high res mesh are ya?
If you are getting all your normal maps from crazy bump / nvidia than you have nothing to worry about.
I didn't mean to come off as saying they where bad artists. I've learned a bunch from them and a trip down the AI pipe (minus the debt) would teach me some good stuff. I guess what I don't like the most about AI's is the false sense of "I'm doing everything right" that can be hard to break thru when you have to re-educate people.
im from art institute of ft lauderdale and there is about 3 people to become succesful so far(that graduated) me working at a small game studio(its ok for now) still working on demo reel stuff... and one at midway and one at nspace...the rest of the 352022350 kids nothing they are lazy...
[/ QUOTE ]
cough 4 , dont forget the other one who works PT with you scuba steve
but yea entire school is made of really lazy people and selected talent, and im saying lightly S-E-L-E-C-T-E-D
As far as overlaying UV's, here's the basics of overlaying a plane (your ground surface for example)
Step 1 - Cut or create your geometry into sections.
Step 2 - In your unwrap, break apart the polygons into individual polygons in the unwrap window. (Control + B is the hotkey for "Break" in the UVW editor)
Step 3 - Overlay these UV's, and place them on your texture in a tillable square. (As shown in figure 1 and 2.)
Basically all you need to do now, is designate a portion of your texture to be whatever that material is going to be, and overlay all of those polygons onto that square. It can be any size you wish, but it has to tile properly for this to work.
Think of it just like you are making a tillable 512 or something, but on a smaller scale and only on a portion of your texture. Using this method some people will model an entire buildings with 1 texture, and designate 1 part of the square for brick, one part for steel, one part for concrete, one part for wood...etc.
You can use this method to really take advantage of limited / small texture space. This method is also used in any sort of game where small texture sizes / uses are imperative. MMOs or handhelds for example.
Hope that makes sense!
thanks for the lil' tutorial. i understand the whole overlapping uvs but i was taught that its not possible if i plan to normal map. however, i did a test overlapping w/ normals with the crazybump program and it worked just fine. so i'm gonna give it a shot with this project. i'm starting to realize just how small 1024x1024 is for this whole house.
The same goes for your spec map...this is where you could possibly also have some strange results in an engine with reflection maps. Play with it and see what you get. Sometimes it works well, other times it can cause issues.
(What up KP and Ben?)
As for the your work. It's looking good. Just put some wicked ass texture work on it.