really need some feedback on final demo...
this is the finished 4 out of 5 environments for my final project for environmental design reel and I really need some feedback before I turn it in
the final pass will have music and ambient noise.
any feedback,suggestions would really help.
https://vimeo.com/65404091
the final pass will have music and ambient noise.
any feedback,suggestions would really help.
https://vimeo.com/65404091
Replies
I think your work in the reel is below average, as far as I can tell all the environments are diffuse textured from photographs off cgtextures, your lightmaps are very low quality and your final scene has zfighting meshes and doesn't even have cast shadows.
You should seriously consider investing time into learning to build proper normal maps for textures either in a sculpting program or in Ddo and making an effort to learn more about UDK's lighting system.
I voted really bad mainly because of the video quality and you asked for votes on the "overall presentation", but I wish I hadn't voted at all looking at the results now. In the future don't make polls, it's too easy to get discouraged by the results
ichll3D "focus on your weakness" what exactly do you mean?
AtticusMars " I don't really like demo reels" every design job requires a demo reel so what are you doing here and why are you criticing me about it?
and then you tell me to stay motivated after there is nothing positive or motivating about any of this.
This shouldn't discourage you. Quite the contrary. The biggest motivation for me as an artist is being better than other artist, and every time I see an awesome model, or scene, It doesn't make me think of how bad i am, it makes me think of what i can do in my work to make it better. And you can't aspire to anything if your the best of the best, so some may envy your position ^^
The only piece of advice I could really share, Sense your focus needs to be so broad, Is to look at Awesome environments (if that is your field) and pick apart what makes them awesome. Analyze and sort out what elements of it make it pure win. Then try to integrate some of those elements into your own work.
Read the posts on some of the major threads (like the ones that roll across the top) and analyze peoples replies. If you don't know what makes something awesome, Maybe someone in the thread does and they point it out. You can learn a lot by studying others work, and others thoughts.
I can learn best by doing. Second best to that, I learn by looking.
The second scene has a couple of decent places, but not much better
The third is a lot better, the street camera was nice, just place more on the ground. Many of your places are really bland. Get a normal for the ground or displacement what not. Remove those horrible stretched normals from your castle.
Redo the mountain its terrible. Id focus on the inside street camera movement and make that as nice as possible.
The last piece, if you had this really strong black shadows like on the main building everywhere, it would look like Sin City and make it really cool, - you need to change it anyways, because its really wrong on all the other buildings. Also place more on the street. Do some small assets like trash.
And to be perfectly honest if you can't spot weaknesses in your work that isn't a good sign... You need to learn to look more critically at your work if that's the case. Compare your work to the work in games you like and think about what makes them different and why.
I'm with AtticusMars. If would help to know your end goal. Because in most places environment art and level design are two very different things.
As for demo reels, if you can show the level well in stills that's always the way to go. A recruiter is much more likely to scroll through several images quickly at his own speed than he is to sit through an entire reel, especially if the first piece doesn't immediately stand out and hold his attention.
As for some specifics, I agree with Atticus yet again,you should be able to see where there are flaws however I'll offer up a couple things to get the ball rolling.
-Your textures seems a bit too stretched or blurred in some places in the first piece (and in others as well). When it comes to organics, perfectly square and uniform UV scale is pretty hard to get, but the closer you can manage the better. The rock texture itself also seems overly noisy
-In the alleyway scene one thing that stood out to me was the harsh transition between the concrete on the bottom half of one of the walls and the bricks on the upper portion. It could be worth the polys to add a bulge, or worth the texture space to add a small strip of mortar between the two to better sell the change in materials.
-The transition between grass patches and the pathway on the castle scene seems very abrupt, try using some grass planes to break up the silhouette by letting them spill over slightly into and over the pathway
-fix the zfighting in the final scene.
I tried to give you at least one per scene, but there's plenty more that can be done better. Take a step back and look at some really strong reference. Look at how similar or identical real world objects are constructed. Observe how stronger game artists pull off similar materials, objects and effects. Study games like Uncharted, Skyrim, COD, Darksiders etc and any game that hits a strong style you want for your pieces, understand what makes their art good, and hold yourself to those standards. If you want to make games worth publishing, you need to be able to show studios that you can make art worth using.
Posting wireframes, texture flats, poly/tri counts etc will help us critique you better. Right now most people are just saying it looks low rez or stretched or bad, because that's all we can really say unless you show us how it's constructed (which btw is what must recruiters I talk to want to see as well) It's also worth noting, for our sake what limitations may or may not be in place for the projects.
And finally, try not to get flustered when you get intense or dismissive critique, the fact that we comment at all means that in some way we're trying to help. If all they can say is it's bad, take a step back and think about why. Look at it with fresh eyes, compare it to what you consider good game art, not good student art, but good art in general. And if you start throwing out attitude, or what looks like attitude, ("nothing useful here... etc") that's gonna draw some crits from people going on the offensive, or worse leave you with no crits at all.
I wrote ya a mini novel there, sorry about the length, but hopefully you get something out of it.
the best way to evaluate your own work is by comparing it to the current standard and currently yours isn't up to the same standard, the second best thing to do is come on here and ask because 99% of these guys will give you the blunt ass truth...
when someone says stay positive don't get discouraged because they didn't praise you, instead look at it in to relation to what you originally asked - you have been given your answer (be it a different one to the one you wanted) its only up from here! (if you choose)
as for areas to look at:
your texture work is flat and you seem to only have a basic understanding of how to texture, I would suggest looking into texturing tutorials and studying how normals/spec/gloss maps can aid in your advancement, also look into uv spacing and uv efficiency.if your going to be looking for an environment artist position you have to make the environments look believable, make them looked lived in and have a logical reason for why they are the way they are, so far most of the areas are too clean and have little believability.
modeling is at the same level, you have a basic grasp of low poly modeling with little/no evidence of knowing any high/low poly work (i.e high low bakes ect) most of the modeling looks like an extruded cube of some sort. I did note a tractor that peaked my interest but it was too dark to see any of it really and scale needs some focus, some things look too large and others too small. also look into modular modeling ect.
the demo reel itself needs work, dont letter box it to 4:3, HD that shit to 720/1080p. also I would suggest either scrapping the lot or choosing one and really focusing your time on a single scene.
the first level reminds me of a pickle and is very very bland so I would drop that as it would need a lot of work. the lighting in the second one is too dark to really show anything off and the purple flash on the lights doesnt make my eyes feel good... logically it doesnt seem to make a lot of sense either. the third level I am assuming is a final fantasy throw back? I would like to see that made in the style of diablo 3/WoW with some nice low poly modeling and decent textures. the last scene I originally thought was your strongest but I would personally re make it in modular chunks for times sake and overall feel/quality.
I think your third scene is your "strongest" piece but I would personally make it again. You are currently at a level where it looks like your going to be needing a lot more practice so you shouldn't sweat it that people view these as under average.
Now I feel I have pointed out the area you need to focus on, I will also point you at some tuts! www.3dmotive.com have most of the things you require help with (they dont pay me for this shit btw... and I no longer have a subscription) they have introductory tuts on high>low baking (for when you are ready) unwrapping and texturing, if you think you are able to match their work to the same standard you are likely to be able to make a scene to the same standard, also the tuition is fine for the most part but its important that you study what they are doing and not just watching the screen... if you assume you can already do the tutorials they have I would say that you are currently wrong, so look into it.
if you cant afford to subscribe or whatever (which you should because its for your education and a future investment :P ) then I would suggest portfolio cloning and before anyone tells me off (I DONT MEAN that you clone someone elses work and pose it off as your own) you should look at other peoples portfolios, judge their level and if they are within reach or slightly above, attempt to make something they have already modeled (again, this is for the sake of studying not for a demo reel or folio). I will go one further and point at ichii3d (not blowing smoke up his ass) go and look at his folio, click university and have a ball! Now if you can model to the same standard as that then you are at a good point but take note that that work is 4-5 years old? (I only know that because I was watching his folio when I too was at uni at the same time - I am not a stalker... ...)
however, dont read everything I have said and get angry/upset, I have spent time trying to not hurt anyones feelings and just said how I see it, its up to you what you do with it.
I wish you the best of luck however and hope to see some wip/study threads here on polycount.
random side note, I used to go to uni with someone who used to post on here and conceptart.org looking for brutal honest truths and he would be delighted when he got slammed by others, he would go off into his room and come back leveled up and then when they would slam him again, he would go back and do the same... that guy is now making games and the likes of me are not... so take away from that what you will.
The overall sense of your scenes had one glaring thing...no scree on the streets. Scree are leaves, garbage, dirt and other things that make the place used by people.
In the castle scene I noticed a mosaic on the wall ( could have been a window)....in a fantasy environment you'd never see this, even in a church.
I felt the film went too quickly to really pick up everything that should be looked at, but over all I would say....
Pick one scene that you like and push the envelope....even better, just take one structure and give it the full treatment with textures, maps, lighting etc., until you've worked through all the issues.
Seems that the projects were too grand to spend a decent amount of time on any of them, this is true of all artists trying to perfect their craft.
For instance, let's say you wanted to model a Dominance War character at 10k tris, but had little modeling experience....you'd be smarter to work through all of the challenges of building a quality 1000 tris model, since more doesn't make something look better, just adds extra highlights as to what you're poor on.
As for feedback...this site is notorious for giving either harsh feedback or no feedback. There are alot of working pro's here and the crits won't be soft like in school....just try to absorb whats being said and see if you can use it to get better.
Good luck with your career, I'm sure with additional time and dedication, you'll be great.
Keep in mind that for most artists, the final 20% of the work will take 80% of the time, but is absolutely essential if you want to stand out.