I am giving up on 3D Modelling :(

Hello Guys!
I wanted to learn 3d Modelling, so I could do everything by myself. I have learned some of the programs for 3D Modelling, like Blender, Maya and so on. I am familiar with almost all the tools and I can do a lot of stuff with these! The thing is... that when I want to make a Model, I start doing it... but in the way, I get confused, bored, and I dont want to continue it anymore, so i just give up! This is driving me crazy! Sometimes the model is not that good for me, because it doesnt have the detail i want, sometimes I dont think that kind of model is appropriate and yea... all the work gets deleted. I dont know what I am doing wrong! I have heard people say that making design in a paper helps a lot, but sincerely, I DONT know!

I would really really really like your Tips guys, coz i am willing to become good at Designing, it just looks like a labyrinth for me!

 

I'm confused about why your title and forum choice are about game design, but your complaints appear to be about 3D modeling.

These are not at all the same thing.

Do you want to become good at designing, or at modeling? And why?

 

Sorry. It was just a mistake of mine! Basically, I want to become good at modelling, because i like it in general, and I would like to make cool stuff using my imagination and programming! Programming is a skill I already own, but 3D Modelling doesnt seem to be the same!

 

I did the same as you. I started many times, quit, started again. Finally I realized that I needed to be able to model, at least a bit to fix problems with purchased models or to make matching items. I stuck with it and now I can model. I did start using Maya rather than Blender, which really confused me. Maybe you just have the wrong tool.

 

Thank you for your tip! But the thing is, I dont use it for editing stuff. I want to create my own. I like to combine programming with modelling. Programming is REALLY Fun for me, and the modelling is also supposed to be, because i have a really extended imagination , But things are not working out.

 

I started modeling to fix models. I now make my own.

One thing I find from people who want to learn something new is that they want to start on top and if they can't, they give up. My own daughters do this at times.

However, you have to start at the bottom and climb that pyramid and defeat that learning curve. You cannot avoid it. If you try to skip ahead and make a fancy automatic weapon, you will become frustrated.

Go find some tutorials of simple things, crates, barrels, houses, etc., and start following them, every step of the way. Take your time, make them as good as you can. If they look bad, start over. Doing this will teach you how to use the tools for your 3d modeling program. You will start working faster and faster. Eventually, it becomes much easier. But...it will be frustrating and tedious for a while.

If you get through the frustration and tedium, you will learn to model. If you give up, jump to making a character before you are ready or a super duper killing machine, you will become frustrated. You didn't learn to read by starting with Shakespeare. You started by learning the sounds of the alphabet and then how to put them together in simple words, like Cat. The boring barrels and crates you are your simple words.

My guess is you will learn quickly if you can use this method and not give up. One thing to remember...start each object from a primitive. So if you want to make a crate, add a cube to your game and start modifying that. A barrel might start with a cylinder. Believe it or not, I didn't start that way with my first model and it was a huge mess. lol

I have learned a lot in the past couple of years. Good luck to you!

 

I'm not sure how long you've been learning it, but for me I became comfortable with modelling when I had a standard workflow and approach. I also suffered a fair bit early on, mainly because I would feel like it was a totally new thing every time and that I had to 'create' the workflow each time. But now I approach things in a fairly standard way, with the same set of steps, and I feel like it takes a huge amount of uncertainty off my shoulders.

For example, it's important to me to model the shape/form first, then add medium-sized details, then smaller details, and I usually approach these steps with the same tools and methods each time. And it should always look balanced, pleasant and complete at every step of the way, even at the beginning when there is just the form.

Also I basically copied workflows and methods off CGCookie videos when I started out and over time adapted them to suit better. I really recommend that site.

 

If you want to combine programming and modeling, why not try something like procedural modeling? Houdini seems like a pretty sweet tool (but it's also one of the most difficult to learn).

Also, are you trying to make models way out of your skill range? If you're just starting out, don't expect to make the next dragon for Game of Thrones. Start with something simple, and when you finish a simple piece, move on to the next one, and try making gradually harder models. You should also leave out normal maps and spec/gloss maps during the beginning, as they just make the whole process so much more complex and time consuming. And don't worry about vertex counts in the beginning, just try to make something.

You might want to try different tools as well. There are more than plenty 3D software out there (3ds Max, Maya, Cinema 4D, MODO, Blender and Lightwave just to name a few) and the tool you're using usually affects the way you work a lot.

Also - I'll be a bit harsh here - you can't expect us to convince you to keep doing 3D modeling. It is time consuming and definitely not easy. If you really want to be good at it, you'll have to do a ton of it. In the beginning your work will most likely be terrible (we've all been there), but once you've made a few projects you'll definitely see improvement.

And as a little side note; if you want to learn more about the theory behind 3D modeling for games I highly suggest checking out the Polycount wiki, as it's one of the best learning resources, as well as the Vertex e-books which provide a ton of great insight about the craft. Video tutorials are a good resource as well.

TAP

 

Just as with everything else, there are no shortcuts. You have to do the hours.
Best advice I can think of is to have patience and keep trying

 

I support quitting in all its forms. Kudos.

 

You need to figure out why exactly you don't enjoy the process. I remember my first steps into 3D modelling - back when I was a child - to be pretty interesting overall. I started out with moving primitives in a quake 1 level editor and I was blown away when I saw how much easier modelling gets, when you use a tool that is actually intended for it (Milkshape was my first). Sure I abandoned lots of projects too because they were too big and I got discouraged by lack of progress. But it sounds a bit like you just don't enjoy the task and learning it at all.
Also... less than optimal choice of forum, "polycount" is the place to be if you want to learn 3D modelling.

 

Of course, you don't have to learn to model. I do not program...I don't like it, and the thought makes my head hurt. I am fortunate enough to have two programmers who do that dirty work for me. I enjoy modeling and so it helped me to get through the difficult parts.

No one says a game developer has to do it all. You either find an artist pal or you spend money on art assets.

 

Is it really supposed to be though? Only a relative few developers actually take on every aspect of development. Most will simply specialize in one area and make up for it by contracting out work, buying up existing assets, etc. I'm the opposite of @Teila, I have no problem with programming but I largely cannot make artwork. It isn't for a lack of trying either.

 

Who wants to do non fun things? Make yourself a good programmer, and team up with a really good artist.

You'll make far more progress specialising in the things that interest you and letting others specialise in the things that interest them.

Edit: Don't give up, outsource.

 

It's fair enough to say, don't do it if you don't like it, but for what it's worth I want to add one thing: transitioning from programming to art and vice versa for me is not pleasant, they clash on some emotional level and I'm usually irritable and annoyed and very unproductive when I switch. I've learned now that I need to acclimatise to them for a while before I can really be productive, and to minimise problems I usually do them in large chunks of time (e.g. a week spent mainly on art/or a week spent mainly in programming).

I know it sounds melodramatic, but it's real for me at least. During art creation I need to be stimulated and emotionally sensitive, I usually spend a lot of time browsing inspirational art and listening to emotional, dramatic music when I start it. On the other hand, for programming I can't concentrate unless I relax and de-stimulate (usually classical music - or none at all).

So if you really don't enjoy it, then yeah, just hire someone. But to give yourself the best chance of success bear in mind that you need to be in the right mindset for the job at hand.

 

I think you might be on to something. Please check out this two-part article:

http://www.modernfilmcomposer.com/2015/06/the-science-of-hacking-creativity-and-motivation-part-1/
http://www.modernfilmcomposer.com/2015/06/the-science-of-hacking-creativity-and-motivation-part-2/

It sounds interesting but I haven't actively tried the method yet. Maybe it gives you an idea where that clash could come from on a neuro-science level.

I feel pretty much the same as you describe, but the thing is, I can't fully switch to either state, I'm permanently trapped between both in my thinking.

 

You know what happens when you get trapped in between? You enter the wretched state of believing that procedural generation is viable!

Wow, I scanned those articles and will read them fully later, they look very practical. I would say I'm a bit like you, in that I'm usually in between. But I haven't really tried to completely fall into the art side of things, might be an interesting experiment.

The article touched on one thing that I think is particularly important: artistic flow requires self-analysis to go away. When you're programming, everything is governed by the rules and logic of efficiency, and you need to follow them in order to reach something that functions the way you want it to - so you need to be always critiquing things by whether they make logical sense. But for art creation, you need to stop thinking this way and sort of 'let yourself go', in the same way for example of you were singing on a stage, or you're at a club trying to *cough* talk to people, or whatever ... the moment you analyse yourself everything becomes slightly nonsensical and it all grinds to a halt. You have to approach it from the simple point of view of emotional gratification, you have to take the leap of faith of guiding yourself by emotional impulses and nothing else, and totally forget about what makes sense.

 

Like any artist would say, just keep doing what you're doing, sometimes, it's not fun but you can make it fun yourself. I do 3d modeling to sell assets and I also draw 2d art. Sometimes you need something to make your work easier, like maybe find the appropriate modeling software/ drawing method that works better for you, then you can actually get to properly model something. It's a come and go process and unfortunately for 3d modeling, there is no fun way to do it, you just have to make it fun yourself and keep working on it! It caught my attention that you said you got bored of it, Maybe it's not what you actually want to do or maybe you just get tired sometimes but don't worry, it's normal!

 

you need to make plans I don't know how good your drawing is, but whatever you can do to make your "goal" clearer for yourself is probably gonna help.
When I started with 3D modelling I always had the impression that I would need to be able to use it like a pencil on paper, no pre work, no references etc. and I was very wrong with that. As long as I wasn't really comfortable with whatever program/technique etc. I used, I couldn't "just" play around and be creative within that program and I learned to accept that I need guidelines (not only for 3d modelling tbh).
so, my advice would be to start planning out what you want to make...in words, in sketches, in kids clay, in references(google) or whatever else that comes to your mind that can conceptualize your idea and then be used as reference for your 3d model.

 

...also don't forget to see your work through the end and don't lose the determination to finish something! Getting something done will always give you a amount of satisfaction.

 

very important lesson I have learned at some point: everything EVERYTHING looks crappy before it's finished
once it's finished though, you can "get to know" your own work and hopefully also learn to like it

 

That's very true, this is still happening no matter how skilled you are, unless you are some pro or genious that can do stuff from a single stroke. When I was doing crappyer work I used to rely on the finishing touches to get something to look good more than I should till the point my work looked like a messy terrible blob.

This is out of the blue but your avatar caught my attention, it seems to say that you draw lol. I am interested in 2d artists and their works no maher what they are.

 

I do draw and i made that avatar myself

 

Make something simple and fun. Like a Rock.
Yes. In game design, rocks are absolute fun and joy to make.
Scale it up big, and you have a giant rocky hill thingy, that your characters could walk/climb up on.
Sometimes when making one object, you end up making, or accidently making something, that could have another
good use for your game. Trust me, it happens alot.
Then when you model and texture that rock, to the best of your ability, make another one.
Or, make another super simple yet fun object, that you would be able to make.

A few good simple examples:
A beach ball.
a plank of wood (in the real world, planks of wood are super boring, but not always in game design.)
a Metroid ball (Yay!!)
a simple, yet cutesy arm cannon, that would fit in well, for a mega man game.

Or, as I posted before, make a simple fun object, that "you" would want to make.

Also, don't forget that patience is needed when modelling.
Sometimes, when trying to make a detailed model, alot of patience is needed, to make it look right.

 

I have done some modelling before. Usually modelling will take more time and patient to make it good. You might just need more practice and patient to make a better model. It is like taking more time to draw a illustration in photoshop. I do not think it is like your programming, where programming can be a straight on answer to do a programming with some logic skills behind it. My old friend used to complain he needs more time to model an animation school assignment to get a A+. You can also research and find some online Youtube modelling tutorial and buy some modelling books.

Here is my video link with my modelling and rigging skill, but just skip and ignore my other animating skills. Hopefully my modelling and rigging skill that took me almost one month to make will be good enough.

 

Also you should know how to model a character properly with the basics to be animatable and riggable afterwards. This is my old modelling notes.

Lol, you will also need to know how to rig a character that is not only about modelling to make a 3D game all by yourself.

 

Right. Gotta take it one step at a time. Here were the steps I took for my Horde Rush hero!
Here's my model sheet:


And here's the whole process:
http://undertheweathersoftware.com/turning-concept-art-into-a-3d-model-in-blender/

 

3D modeling is not a matter of design, but of artistic expression. It is like illustrating, or sculpting, or painting. In fact, it incorporates elements from all of those disciplines. Arguably, you should know how to draw, and sculpt, and paint before you start trying to create 3D models.

But not everyone dives into this sort of thing at the same point. So it is understandable that you might feel frustrated starting out. Here are a few basic pointers.

1. Polycounts: 3D modeling is essentially made up of tiny triangles. No matter what program you are running, this is what it ultimately boils down to, tiny triangles. In 3D modeling lingo, they are called polygons. The number of polygons in your model determines how complex the model is. It can also affect the performance of your model in a game engine. (such as Unity) If performance is important, than reducing your polycount is a good way to cut back and have your game perform a little better. With modern GPUs, polycounts aren't as much as a bottleneck, so you usually don't have to worry about them unless you are developing for a mobile platform. But it is still a good idea to keep an eye on your polygon density, and make sure that the right level of detail is being applied to the correct areas of your model.

2. Silhouette: This is a fundamental element of any 3D model. It is important to look at the edges of your model, to be able to see what kind of silhouette it creates, and how that silhouette changes from different angles. This will reveal a lot of artistic basics. Good models will often have strong, recognizable silhouettes that look good from a number of different angles. It is a good idea to make sure that your models have asymmetric silhouettes. (one side is at least slightly different from the other) While pure symmetry is easier in modelling, it is less visually interesting. I often start off modeling a symmetrical model but shift over to asymmetry when it comes time for some of the more detailed areas. A good way to inspect your models silhouette is to apply an all-black material to it that ignores lighting. Then just rotate the view around and look at it from different angles.

3. Texturing: UV mapping is a fundamental of 3D modeling. It can also be quite challenging, especially for someone who is not a painter. I'm a solid modeler, but have never been a particularly good painter, and texture mapping is one of my weaknesses. Thankfully, there are a lot of much better tools these days for creating UV maps quickly and easily. You will still have to handle the texture painting yourself, but stretching out the UV map is a simpler process than it used to be. If you, like me, are a bit weak when it comes to painting, you might consider an art style that relies less heavily on detailed texture painting.

4. Digital Sculpting: These days, there are now tools that allow you to "sculpt" models instead of constructing them from pure points and polygons. This is often a more comfortable approach for some people, and is also one of the better ways for creating extremely detailed 3D models. Most 3D packages these days feature some manner of sculpting tools. Even Blender provides a basic sculpting solution. If you find that this approach appeals to you, you might consider doing your first-pass modeling in a sculpt-focused program like Z-Brush or Sculptris.

 

If you're fortunate enough to own a Ps4 and two motion controllers then the game DREAMS is in the works by Mm and there you can model stuff with motion controls which should be faster, easier and you can even export it as .obj

 

Everyone has different mindsets. There are some fantastic modelers out there that can't code. But you can. I mean, it takes alot of practice. For them to also code would take years to become just apprentices with any language. But my advice is to just keep practicing. I've been into the idea of 3D modeling so I can just quit googling all night looking for a fine bowling ball for free... But it is definitely a hard and long road. Well, I wish you luck if you decide to keep going with modeling!

 

I find sculpting way more fun and easier than modelling,
I can get everything right where I wanted, maybe you should give it a try!
But then again, retopology is a huge pain..
Unless you want an asset that is not fit to be used for game that is!

As everyone has said, try to get a workflow.
I am currently interested in learning hard ops workflow to create cool inorganic objects.
Taking a break and learning new workflows is always very fun when one thing does not go so well!

And oh, texture painting is also damn fun if your PC isn't a potato like mine

 

Did you know we can determine whether someone will good at math, by asking a few, simple questions? This research led to the discovery that grit is the best predictor of success in math (not IQ). Grit is the skill to persist in the face of adversity, set backs, and time. I will offer that being a 3D modeller, a coder, and a designer, also follows this truth. If you want to be one, you will put in the YEARS of hard work, persisting to get good.

TL;DR - Try, Improve, Repeat.

Gigi

 

The key is to do something start to finish. Don't stop in the middle. Finish your model, unwrap, texture, rig and animate it. You'll learn a lot from one model. Start with a simple cartoon character if that is easier, then move to more complex designs.

When you get done with it take lessons learned and apply them to the next model. Learn from your own mistakes and improve on your weaknesses. Try speed modeling. Watch speed modeling videos and try to keep up with them.

Speed modeling videos actually helped me a lot when I was mesh modeling, I got to the point where I could do a model that originally took me a week in about 2 hours just from modeling something super quickly from start to finish repeatedly over and over. By unwrapping, texturing and rigging you'll also expose a lot of edge looping and polygon distribution flaws that you'll fix on the next model.

Bottom line - suck it up and learn to be better instead of settling on surrender.

 

Study meshes, polyflow, polycount needed to achieve different looks. After some time, you look at (real) objects and you start thinking '3D', imagine how you would model all those different shapes.

When polymodeling I just use very few tools/modifiers in max, like chamfering, mirror, bend...

Imagine you are an ant, walking over different shapes. Most car bodies (and other stuff) for example are bezier curv-ish, something interesting to look at (Google)when learning how things are shaped
And your tools are an infinite number of triangles/quads (polys) to match those shapes. It's an abstract topic, just like programming is.

Start with non-organic stuff!
Go to smcars.net and search for blueprints of a simple car. Try to find Tamiya(!) blueprints as those are the best ones.

I started modeling 15 years ago and I still remember, how I wanted to give up, how stupid everything I've done looked. Practice!!! Modeling is like drawing, it IS a learning process. It will take you ~1 year to get a solid workflow, i'd say.

 

I'm only tangentially answering the thread, but there is this idea than code and art are somewhat magically separated and that the process is different ... Art still ask for analytical skills, but the kool aid is telling art must be the kind of brain fart you magically get right on a single stroke of genius. Let me show an old drawing, not the best but rather illustrative.



On left I had no idea what I was going for, I just wanted to mix iconic element of popular characters. On the left I refactored it as an idea come to my mind (hero as fallen, and the sidekick must take is place, which is reluctant to do). It looks much better much more focus, it conveys something, while the other one is stiff and more like a collection of part, it tells nothing. Understanding what you do and what you want guide you. It's notable that the second part took me less time than the first too (which when you are stuck within it make you believe you are useless).

Generally beginner get stuck into believing they have a single shoot, but artist has many shoot at it, just like code they refactor the art until it hit their target. Just like code you have to know basic architecture and practice to master it.

Recently I have start modeling by eschewing sketch first, I do a crappy model with bad topology and juxtaposed volume, then refactor (retopology) once I find the correct way to do it. It's fun! I get to play directly in the space!

Put the whee! before the cart!

 

Isn't the "blockout" phase in 3D modelling a common practice amongst professional modelers? Get the big volumes down first and see if the overall design works before you waste time detailing a dead-end concept?

 

I was kinda where you were when I started 3D especially on things like airplanes and cars I just could not get the models to look correct at least to me they didn't. Other than those I have had it easy in the 3D modeling programs just because I think I can see things at a different angle. When I look at a object on a flat piece of paper or in a reference imagine I try to imagine as hard as I can what that object would look like of it were right in front of me. This works with small to medium objects like furniture or toys, however this does not work to well for larger object with much more detail, like aircraft or even yachts. There to big for me to picture in front of me so I just keep trying different angles and eventually I would find something that would work like instead of starting with a cube lile most models do, I would start with a sphere and boom an airplane would appear. This helped me a lot so hopefully it helps you some. It also helps if you have plenty of references images.

 

Go simple. You are suffering from the most common problem facing new artists in any medium. You dream big, but frankly... you suck.

Don’t worry about that though. Everybody sucked once. That’s why you start so simple. Find something fun that you want to build and just do it. Don’t worry about using it in a game, just learn.

 

Necro'd from the depths.

 

Would be interesting to see if he became a great 3d modeller in that time though haha

 

3D modeling is just like any other skill. You aren't going to produce a really high quality product until you've put your 10,000 hours of practice into it.

 

I think a problem a lot of beginners face it that, they want to work on games because they love playing games. But games train you to be impatient, and offer satisfaction for minimal effort. Especially if you play games a LOT, this does effect your thought patterns.

So then you jump into making games, or 3d modeling or any other part of the process, but this is nothing like playing games. It's real work. You do a lot of tedious labor, a lot of hard figuring, and where is the reward? Nowhere in sight.

So you've got to change the way you think, and learn to enjoy work -- something I think a lot of people fail to appreciate throughout their lives, and it's a shame because you spend most of your time doing it.

 

And then you need to think like a player when designing and testing ... making games definitely needs a lot of changing gears.

 

Right. You have to understand yourself, so you can analyze what is really happening in your own mind while you are playing games, so that you can extrapolate your experience onto others, and of course understand the limitations of this...

Know thyself, yada yada yada.

But yeah, you play a lot of AAA games recently, and you wonder, "do these guys play games at all?" And the answer is no. Not the people making the big decisions anyway. That's why mainstream games look more and more like gambling.

 

On the flip side, game development has sort of directed my gaming tastes towards turn based strategy games, away from FPS and RTS games when I was younger. Or maybe I'm just old now.

 

After about two nice long years, by now, you must be a good 3d modeller.

 

I found that doing some tutorials help. Just going really slowly through them step by step.

Also, do you have a sketch set as your background? That makes it about a thousand times easier!

In fact, I prefer 3D modelling to 2D. Because in 2D you have to do all your own shadows/shading/perspective etc. So 2D is really more difficult!

 

Hey, I do that too.

 

I suck at modelling. That is why I keep trying. First I tell myself modelling a human isn't something I should try yet, because humans are complex. Even cartoony ones.

But I keep doing it, stubbornly refusing to give up. And guess what? After 3 months of dicking around in Blender, I succeeded in pulling a mesh over Blenders default metarig and make myself a humanoid avatar. It took some time and I watched a lot of tutorials on general Blender techniques.

It's nowhere perfect, it has silly stick feet and a ball for a head. It has separate meshes that intersect. But it's a start, and I can make it run around in Unity. It even ragdolls nicely.

What worked for me was to worry 100% on the technicalities and workflow, and not care about visual style and flair too much. I understand the desire to use cumstom made models and really own ones work. For this purpose, I've decided to build a game / story around the model instead of vice versa. I found it a nice approach for my first attempts at modelling a humanoid.

 

@Inxentas
Nice story.