[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Search] [Home]
Board
Settings Home
/3/ - 3DCG



Thread archived.
You cannot reply anymore.




File: PleaseHelpMe.jpg (45 KB, 549x370)
45 KB
45 KB JPG
Evening /3/ I'm here to ask for guidance.

I just got home from a 12 hour shift at a shit job. I need a hobby and honestly I'm just tired of this work-sleep-work 90% of the time.

I have an interest in various things, chief among them being some sort of creative outlet. I have as much imagination as the next guy but I figure thats useless unless I figure out how to get those ideas out of my head.

I would like to learn 3dmodeling, I plan to use Blender as I don't exactly have a ton of disposeable income. I can however set some aside for proper training/learning. I figure the best course of action would be to dip my feet in for a month or six. See if I have any aptitude for it.

Im assuming blender would be more than enough for a hobbyist like myself. However I would love to know if theres a /3/certified "path" I suppose to learn 3d modeling. As I type this I'm reading the sticky but I have a couple questions that as far as I can tell aren't really covered.
>>
>>648098
Long term goals: (5-6 years)

I would like to make a game, I have no idea on how to approach that. I have no grand dreams of being the next minecraft or pixar. I just want to be a guy who makes some cool stuff. I picture in my mind a simple tower defense game or maybe a 5-10 level platformer, nothing crazy.

As part of making this game I would like to do everything, art and technical side. However if I'm honest with myself I have a tendency to overwhelm myself. Pursuing both coding and artistic development would probably lead me to procrastinate and eventually abandon everything.

As such a game would be nice, but a movie would be ideal. I say movie but what I mean is a short clip, something similar in length and scope to the "introducing" series for TF2 or a WoW expansion intro video.

Short term goals: 6months - 2 years.

Basic understanding of modeling, texturing, rigging and animating.

My question is on how to best approach the world of 3d, while I'm superficially familiar with some of the terminology and the like I'm firmly stuck in the "I know enough to know I don't know much" stage.
>>
>>648098
>>648099
Example issues:

1. When learning modeling I realize there are different requirements/limitations for still models and those that are to be animated. Would learning to model "stills" and then adapting to model with intent to animate be a duplication of efforts? Would I learn bad habits or otherwise shoot myself in the foot down the line?

2. As far as modeling vs sculpting, am I ok to ignore sculpting for now? While my hardware is (I assume) fine (Ryzen 1700 + 7970) I don't have a tablet or the like. I also can't draw, which I realize is a disaadvantage compared to many artists.

3. Should I hold off on 3d for a year or two and pick up drawing? Would drawing during my lunch breaks at work and commute be beneficial?

4. Is there a path for me to accomplish my goals? I have noidea if even my timeframes are reasonable, I know I can dedicate .5-2 hours a week to this, maybe more on days when I can't pick up overtime. I just feel lost on how to go about getting started. I assume this is the value in going to school for this? In an ideal world someone can point me in the direction of a good free resource (As far as I can tell blender guru is my best bet?) for couple months before I go onto some kind of paid tutorial service? Or would I be better served by skipping the free stuff completely?

5. As financially blender is the safest option, I have concerns about the 2.8 redesign of the UI, since I am just a beginner I assume I would be better served by starting now and understanding fundamentals and dealing with 2.8 when it comes out (any idea when that would be?) Part of me wants to wait but I hope someone can offer some reassurance that 2.79 would be fine to use even after 2.8 releases?
I'll be around for maybe a couple hours now, but will happily check thread from work in the following days as I know /3/ is a slow board. Please feel free to ask any questions for myself or other hopefulls that might be keeping an eye on the thread.
>>
>>648099
>My question is on how to best approach the world of 3d, while I'm superficially familiar with some of the terminology and the like I'm firmly stuck in the "I know enough to know I don't know much" stage.

you really don't need to learn any "term" conventionally. but if you really want to plan ahead learn about topology, polygons,vertices's,3d space,extrude,move,bevel,shaders etc.
usually a few introductory videos will get you up to speed about everything, blender have alot of beginner guides on youtube, and most of them are decent.

>When learning modeling I realize there are different requirements/limitations for still models and those that are to be animated. Would learning to model "stills" and then adapting to model with intent to animate be a duplication of efforts? Would I learn bad habits or otherwise shoot myself in the foot down the line?

when you are a beginner its best you learn how to make animation ready models first, with very low polycounts (somewhere in the 3k to 5k at most). topology is your enemy and friend. you hate it but will need it.

>As far as modeling vs sculpting, am I ok to ignore sculpting for now?
ignore sculpting
>Is there a path for me to accomplish my goals?

seems like your timeframe is more than enough to become decent at 3D. remember, you are supposed to have fun with this. forget about the technical stuff and just enjoy yourself.
once you get up to speed, you can become more technical with your models.
you can be served by free stuff completely, the level of free stuff on youtube has gone up alot. some of it is even pro level if you look in the right places.
>last question
think of blender 2.79 as the most stable version of blender, and think about 2.8 as a very pimped version of blender with some bells and whistles. your work won't fall short if you just use 2.79
>>
>>648119
>when you are a beginner its best you learn how to make animation ready models first, with very low polycounts (somewhere in the 3k to 5k at most). topology is your enemy and friend. you hate it but will need it.

OP, ignore this part.

This shit meme needs to stop.
Characters can reach and go over 40k quads (80k tris) before shit starts to slow down noticeably.
You can plop that shit right in-game and it will work smoothly.
That high polycount is obviously reserved only for the main character. Don't try to spawn a crowd of background npcs each with that many polys or the engine is going to spit in your face.

LoL characters are 10k tris each and they're seen from fucking orbit, and the game needs to run as smooth as possible because muh competitive dick measuring contest and muh latency.

If you're rendering, then it literally does not matter.
In movies, the character mesh is cut at the joints so that the animators can work with a rig that runs super fast but looks like calamari. Then the mesh gets swapped down the pipeline with increasingly higher resolution meshes, up to 700k+ tris for the one that gets fed to the render farm.
And all are derived from the same high res sculpt.

If you start telling beginners that models that need to move can't go over 5k polys (tris or quads?) then they'll be left wondering how we ever advanced from the ps2 era.

>topology is your enemy
Topology is your friend.
Only 'tards hate topology.
Start with edgeloops at every joint, then connect them. 8 sides. Keep as low as possible, increase only when needed. If a silhouette-defining detail gets lost because of low polycount, add another edgeloop.
Don't stress too much about it. If you're animating it, nobody will care nor notice minor flaws. If you don't, nobody will care nor notice minor flaws.

And for the love of God, start drawing.
Learn form and anatomy and everything will make even more sense.
>>
>>648100
1. if it's modeling anything that is going to deform in future it is best to learn modeling them animation ready. It may be hard for you what it means until you start learning rigging, but there are many resources on learning good topology. You shouldn't be starting with character models anyway.

2. You should learn sculpting when you get to stage of making detailed character models. If you want to make your own characters then you should be studying drawing anyway. You don't need sculpting for making cups, plates, swords and other props you are going to be modeling as a beginner, so stick to modeling for now.

3. Drawing is going to be beneficial, but no reason not to start 3d until you get good at drawing. You can do both, and you still don't even know if 3d is good for you.

4. I cannot tell you which blender youtubers are the best. I can tell you thou, youtube is full of good free tutorials (often better than what you would pay for), there is many that are made by hobbyists who teach you bad practices, or techniques that are outdated. Unfortunately same goes for paid tutorials, and as a beginner you have no way to tell. If you want to stick with blender, then check out their general threads and ask there.

5. The UI redesign is actually to help beginners, since you are going to find out that blender 2.79 and below has a pretty unique UI that is hard to start with. Sure getting familiar with your chosen software is important, but you should be learning techniques and concepts, so you could essentially use that knowledge in any software.

>>648099
Your short term goal is reasonable

Your long term goal of making a simple game is also achievable.
You gotta understand there is a difference between making game assets and programming the game. If you want to make everything yourself, you will need to learn everything. Same goes for making a character, and animation.

Start with really small projects that are achievable in shorter time.
>>
>>648128
This advice ONLY makes sense if you are willing to do the work to generate a split-geometry animation rig and if you truly don't give a fuck about seeing your deformations in realtime, and you want to write some sort of pipeline for yourself to toggle between models.

That is one possibility. The other is to prioritize good topology, keep your polycount reasonable and make your rig fast and fun to work with at full detail. It's not a meme that needs to die, it's a better, more pleasant experience for beginners.
>>
>>648171
>This advice ONLY makes sense if you are willing to do the work to generate a split-geometry animation rig and if you truly don't give a fuck about seeing your deformations in realtime, and you want to write some sort of pipeline for yourself to toggle between models.
Which, by the way, is much easier than you make it out to be.
You just need to not be a lazy fuck and not be afraid to learn and get your hands dirty.
Which is something I learned with time very few people are actually willing to do.

>The other is to prioritize good topology, keep your polycount reasonable and make your rig fast and fun to work with at full detail
The whole point of my post was to say that you don't need to get stuck with ps2 era polycount if you want to see characters move. 60k+ tris are PLENTY to put all the detail you want and still have a fast rig. Which is 10x the initial advice.
>>
>>648098

When I started to learn 3D in general I also went the route of learning to make my own models first. Getting the first acceptable mesh down took me like 2-3 weeks, (and by acceptable I mean by my standards back then, so dont picture anything good, it just worked as a character). Then I learned basic rigging from the digital tutors courses back then so I could rig the thing. Then animation. Then rendering. I just did the whole pipeline from start to end. Once I had it all down, I reiterated everything again. Improved my models based on my experience with rigging them, improved my rigs based on my experience animating them, improved my animations based on the renders/playblasts I made, etc. etc.
I went with second passes, third passes, always making new better characters and rigs because I constantly kept "upping" my standards to work towards a new sub-goal of quality.
I think thats an important aspect that I see many novice people fail nowdays. They want too much too soon. It took me an entire year to be at a level where I knew the basics of all those fields. But that was just the beginning. I realized quickly that my biggest ambitions lay in modeling and rigging so thats where I put all my focus on. An other 2 years later I got hired as a rigger by a big vfx studio. Funny thing is that when I entered that job and saw how things were done in the big industry and how little, amateurish and puny my own work seemed in comparison really humbled me and I felt like the beginner I was 3 years ago. Thats when the true learning began..
Anyway.

tl:dr:
Set yourself small goals. Recognize whats within your ability to make, and do it. Then improve based on what you did. Then again and again and again.
>>
Hui
>>
Op here, I apologize for the delay, and appreciate the responses. Just a couple followups.

From your posts I gather that while I should keep in mind and learn the differenced between modeling for stills and animation it might be worth delaying that a bit until im comfortable with blender and have finished a few turorials and actually have something resembling experience.

As far as sculpting Im guessing its relatively important for my development but can likewise be delayed. I guess my plan right now would be to draw in my down time at work (and i guess apart from "practice a ton" ill find some kind of resource on that) since it would help for understanding topology and proportions and the like for future projects. I imagine itll be frustrating but so is learning anything I guess.

Am I right to assume that both topology and poly counts are that I should not obsess about. In the sense that hitting certain polycount targets or perfect topology are a waste of time? Im guessing I shouldnt aim for more/less polys or obsess about topology until I understand why it might be good/bad and instead would get better roi on iteraring and completing various smaller projects?

A tangential question, is there any resource on how to organize all the files Ill need for 3d? Perfectly in a portable manner? I tend to be a pack rat and having some sort of guidanxe would be useful for say being able to copy a directory and render on another computer? With semi-constant hardware changes I fear I might lose all my work from a year ago for instance.

On the subject of the game, as far as I can tell unity seems to have a ton of turorials related to blender, would that be a fine option for a hobbyist like me?

Im gonna go ahead and start with blenderguru and look into borncg whos name Ive picked up somewhere. After Ive exhausted 5ish free tutorials and see if this is for me, ill look at paid options.

Anything else I might be missing here?
>>
>>648172
>The whole point of my post was to say that you don't need to get stuck with ps2 era polycount if you want to see characters move. 60k+ tris are PLENTY to put all the detail you want and still have a fast rig. Which is 10x the initial advice.

Ironically you can still do a shitload of stuff with PS2 / mobile tier graphics and if you can demonstrate your ability within that limitation you can still get a job. My biggest mistake the past couple of years was focusing on HURRR FILM pre-rendered graphics tier shit and totally ignoring mobile quality. I missed out on sooo many jobs and now I'm out of the industry completely, fighting to get in (to mobile)





Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.