I obsess about accuracy and produce next to nothing.

Please forgive this introspective ramble spilling out into the forum, but it’s something I wanted to get off my chest. I have been interested in Blender for almost a decade, have kept up to speed with the UI changes from 2.3x to present and get on fine with Cycles. The problem is I never finish anything, or complete it to the point where the results are worth sharing. “Blender and me” these days is little more than watch a tutorial, try it for myself, go “Hey, that’s nice! Blender rocks.” and close it down.

Andrew Price made a really nice podcast about motivation and resistance, but the obstacles I face didn’t come up unfortunately. In a nutshell:

  • I insist on modelling everything myself. To call it my work, I have to have pushed every vertex with my own mouse. Call it mega-OCD and impractical if you will, maybe that’s the root of the whole problem.
  • I obsess about accuracy. I can’t just model a coffee mug, it has to be the archetypal coffee mug that I have painstakingly captured in 3D and can potentially re-use.
  • I fear off-model objects and bad topology. If I can’t measure something accurately enough (I’m not prepared to cut that coffee mug in half) or think my mesh is becoming overly complex, I lose interest.
  • I have a family and can’t justify hours of Blender time. My creativity doesn’t get many chances for an outing, and when it does it is usually crushed by the above.

I know that Blender is an artist’s tool, and having a rod up my ass about an off-model table leg is probably the worst approach. I’ll never give up with Blender, but it kills me to think about what I could potentially do when all I actually do is watch Blender Guru and explode / melt / grow hair from the default cube, laughing my ass off.

Yeah, I’m sure a lot of people would say you don’t have to produce anything with Blender to have fun with it, but it feels like a waste. I love this program and know how to use it but… nothing. Has anyone else felt this way and does anyone have any advice?

Thanks for reading!

There are some who say that someone who obsesses over perfection to this type of extent may be someone with a mental issue, it shouldn’t really matter if the measurements are fully accurate as long as the vast majority see the final result as visually acceptable.

For example, don’t bother if the subdivision surface isn’t of perfect roundness as full zoom, you will not notice a thing in the final image. It’s the same with textured objects where a millimeter or so of stretching will not be visible. If this obsessive perfectionism is getting to the point where you can’t function normally, then you might need to allow your family to help give you some reflection on your thoughts or make an appointment to see a professional.

Try to think different for one time. Don’t decide to ‘model’ something, rather decide to create an image, visualize it in your mind, then start to set up what is needed to achieve your goal.
Point to the result only, cheat where you can, all is allowed.

Play a new game.

paolo

Man, I say just keep doing what you’ve been doing and find a way to leverage it. It is much easier to transition from being overly detailed than it is to transition from being not detailed enough.

But do try to find your stumbling blocks. I have similar problems, where I spiral into a quagmire of details. I try to start broad and then add in more details later, but I start zooming further and further in. All of a sudden, I am modeling sub-pixel detail that will never be seen! The best thing that I did was to pick a project that I could really get into. I modeled a handgun. but not just like a game ready gun, I modeled every screw, every ridge of the diamond grips, every bit of detail I could from the blueprints I found. because it was a series of parts, I was able to complete a part at a time, which helped to motivate me.

I really feel for you man, it can be frustrating to do so much and have so little to show for it. But attention to detail is an asset. It can make your work stand out. And practicing good topological habits can really pay off too. Sometimes I take a shortcut to save time, and it burns me later when I need to make a change, but my topology is hosed.

The best way that I have found to verbalize this conflict is summed up in this sentence: “There is no such thing as cheating, only less robust solutions.” If you want to make the most robust solution, it’s gonna take some time, but if it’s worth it to you, it’s worth it. But, don’t be afraid to bend the rules a bit, if you want to speed up your process.

QFT, sourvinos is right.

In music they say “if it sounds good, it is good” and they’re right. If it looks good, it is good.

You might try Eno’s oblique strategies, or simply giving yourself rules to work against - this is considered (and is, in my own experience also) a great way to stimulate creativity.

“Make a piece using only 400 faces”

“make a piece about a day in your life using only diffuse materials and cubes”

“Make a piece about your mother where at least half of the image is occupied by a noise texture”

and so on.

Decide what you want to produce before even approaching the computer. If you feel it’s too broad, give yourself a rule or principle to adhere to while creating the piece. Then, don’t start anything else until that one is done.

Remember, if it looks good it is good! Forget technique when technique gets in your way. Techniques are simply means to an end, and using “good” technique but accomplishing nothing is infinitely worse than using “bad” technique to finish a piece you’re overall happy with.

Another thought: 99.99% of the very best artists don’t use CG at all, let alone Blender. So clearly, good technique with Blender or 3D tools is not a requirement to create good art.

I think to create truly good art, you do need strong technique, but a technician is not the same thing as an artist.

Another alternate thought: maybe you would have more fun working on 3D prints and things of that nature. You say you aren’t interested if you can’t measure something accurately. What does that have to do with art? Devil’s advocate: maybe you’d have more fun on the ‘engineering’ side of things… just a thought.

To add to the “create an image” idea, it might be worthwhile to layout an image using a “blocking in” technique only using primitives or extremely simple objects. Create your composition, even go as far as lighting it, now you’ll have a better idea of what you NEED to make the final image.

Good Luck:)

The following may not be good advice for quality art production but may help with changing strategy.

1.-Make a hard deadline.
2.-Stick to it.

That’s it. If you practice breaking down the project into chunks of time remaining you may be able to stay in touch with the scope of a project instead of the minutiae.

Thank you all for the replies; some really insightful stuff here. I don’t think I’m crazy (not “I have to flick the lights on and off 15 times when I leave the room or my family will die” crazy anyway) but I have always been a perfectionist and annoyingly pedantic. My chosen career probably doesn’t help either - I’m a Unix sys. admin!

Case in point, I just had to stop myself manually replacing the hyphen above with an en-dash (ALT+0151) ARGH!

I think Kemmler makes a good point; this is something I have been conscious of for a while. The CG world is where art and technology collide and I am a techie first and an artist second. That’s why most of my renders are of models alone rather than scenes. I need to stop thinking about the desired end result as “look how good my model is” rather “here is a picture”.

I did the same thing; I play guitar and decided to model a Fender Stratocaster. The Strat is an icon, instantly recognisable and quite popular with CG artists but I have seen more bad models than good. Mine would have to be spot on, or there would be no point in doing it other than as a learning exercise. I had big ideas of photo-realistic renders of it with various paint jobs, which I could share with other guitarists.

The Strat body is an organic but very clean shape, like a car (as opposed to the Telecaster which is practically 2D extruded with a slight bevel). It has contours at all angles and directions but if any part of it were misplaced or too boxy it would look awful. I spent ages with it on my lap thinking “how am I going to do this?”. A lot of work with subdivision surfaces and creases later and I had something I was happy with, even though it wasn’t 100% accurate. It needed more work but I started modelling the neck and headstock (easier but still quite tough to do).

I then got an idea of how much work it would take to finish, so I sort of left it. I still want to finish it, but I don’t know when that’ll be. My face is gaining geometry at a faster rate than my models.

I’ve never seen it done right. They never get the details how they should be. I’ve often thought to try it myself.

I know what you mean man. A couple of years ago I was so obsessed with into microscopical level detail correct edge loops that I totally dissed all images that didn’t have good animation and displacement topology. The uv-maps and the whole scene with naming conventions and organized hierarchy should be flawless. I was totally obsessed that everything should be perfect that I lost track of my spontaneous creativity.

Today I look at the bigger picture and enjoy my work so much more. I occasionally still think in terms of perfection but I let it go and move on. I allow myself to flow.

This is a funny topic…

Some suggestions…

  • Make a small game (BGE/unity… whatever) - give yourself a target of ~40fps to maintain. Its normally a pretty good constraint so you don’t get carried away with detail and focus on budgeting detail to achieve efficiency.

  • You COULD try use a CAD application, I was talking with the developers of BRL-CAD, where its common to model every detail of an object - then do physics or raytraced rendering. (No doubt other CAD apps work the in a similar way). - So if you like to indulge in detail maybe some other application is better suited to your needs.
    However I wouldn’t do this kind of work for the sake of it, there would need to be some purpose to make such detail.

I do want to stick with Blender. I know the software is more than capable of the accuracy I desire, I’m just not sure I am. I don’t want to spend a whole week modelling a guitar, but I would rather do that than rush it and never be happy with the result. That’s why I produce nothing!

To give you an idea, this took me several hours, and it’s still not quite right although I am happy with the geometry so far (yes, those are all quads):


The problem is I did this months ago. It was my last project before I hit the wall as described by my original post! I was thinking “Hey, I should really model the cut-outs for the pick-ups, tremolo arm and bridge springs, then I could do exploded diagrams and stuff!” then I realised how long that would take and stopped. Since then all I’ve done is play around with Cycles and dream about doing precision modelling for a living.

Yeah, funny thread and topic, also considering that I think ?'m prone to the same crazyness.
Luckily, being using blender fulltime at work now, and with un-realistic deadlines, i HAVE TO give up on too tiny details and carry on. Surprisingly I’ve discovered that often it makes no-difference if I spend time on super detailed model&texture, or if I just take an easy path (box-ish object with a stitched texture ;))
So, my suggestion is: work in a layered fashion. I.e. if you want to make a scene for a still, get the whole scene done roughly, place lights and render. Look at what is off and refine. Reiterate the process until you’re happy. Likely you’ll be surprised that many things will look good, even if the little Gemini Cricket inside you is screaming “Mooore detaaaail!”.

One more important thing: being not your job, the problem might be motivation. Give a try to contests. Deadlines are your friends!

So that is the problem!! :wink:

/me nods … :RocknRoll:

I think that we’ve all done that … and it’s not a “mental issue.” (Except to the extent that ‘playing with Blender’ is … :stuck_out_tongue: )

The trick, I think, is how to balance that obsession-with-detail, with the pragmatic need to get a project finished and into the Gallery, or at least into your bank account. You need to be able to choose where-and-when you should “obsess with detail,” vs. when it will add nothing-at-all to your project.

A good way to achieve that is with “stepwise refinement.” Start with “the whole show,” using (initially …) simple geometric shapes that are of the correct scale size. Figure out where your cameras are going to be, how they’re going to move, and how you’re going to light the set. This will also tell you exactly how much of the set, and of each object, you actually need:

  • If you can’t see it, don’t do it.
  • If you can’t see it well, don’t do it well.

You’ll naturally tend to produce stuff that you’ll never use … to pour “blood, sweat and tears” into it in case you “might” use it. That’s inefficient.

Hollywood never built the back side of a movie-set. If you look at some of the existing models, you can see exactly where the camera’s view started and ended, and so you knew that the set-designer knew that (“down to the inch”) at the time.

Models were also built to many different sizes and levels-of-detail: if the starship was far-away or moving at warp speed, it didn’t need detail. Star Wars Episode One actually shipped to theaters with Q-Tips cotton swabs, painted different colors, representing a crowd in the podracer sequence. (Since your eye had seen a previous shot of real actors, it “saw” them again in a subsequent passing-shot. The eye can sometimes easily be fooled. It can also be directed.)

Obsess, by all means, on things that need to be obsessed over. But, nothing that doesn’t. Don’t obsess over details that “you think” “you might” use. Instead, determine in advance what will be required, and use the technique of stepwise refinement to zero-in on the final with a minimum of wasted effort. You’ll probably discover along the way that the set of things for which obsessiveness actually results in payoff, are completely different from what you originally anticipated.

This will also help you see “the show as a whole.” It will make it seem much more finished, or nearly-finished, which helps a lot with the grind.

Danny Darko, What you’ve described IS an obsession rather than simply a passion. Understand People can become obsessed with anything, work, golf, gambling, drinking, cooking, CLEANING, or even creative outlets like 3d work. Why I say this seems obsession rather than a creative outlet is your reflection towards family. That was a cue. Your time with Blender is stealing time from your family. Irreplaceable time by the way.

Work should never steal time from one’s priorities. And Hobbies can become the worst time suckers. I’ve experienced a similar problem with Blender. In fact, I just deleted a 6 GB sized file of a project I’d been working on for several years just because of the same issues. I did it because you’ll survive without it. And in the end if its meant to be, you’ll return to the work. Moreover, when and if you do return, you’ll surprisingly discover your efforts will likely have improved, tangibly. Don’t ask me why, but there is usually a noted increase in ability after a dedicated lapse - kind of an artistic given.

In the 1650’s St Mary of Agreda worked ten years writing the Mystical city of God. She had wrote thousands of pages of parchment and it was nearly complete when a fire broke out in the convent and consumed her work. But undeterred, after a lapse, she returned to re-writing the work and this time finding her writing was better written, more readable. And not long after it was published and is still in copy today.

I read an article once by an author who had a similar problem.  He loved to write.  He was always writing, writing, writing.  But nothing was being published.   His family harped over his lack of interaction.  Finally he listened to their concerns and resolved to devote more time to his family.     Yet He continued to write, however, only at night after everyone was asleep.   He attended to chores and family activities during waking hours.

  Sure he was tired the next day, but he didn't neglect his family anymore.   And the result was that his writing actually became better, his life more disciplined.    And stuff got published. 

But aside from the obsession, are you an artist? I think you might be. Consider this quote from STeven Pressfield … “If you find yourself asking yourself (and your friends), “Am I really a writer? Am I really an artist?” chances are you are. The counterfeit innovator is wildly self-confident. The real one is scared to death.”
Steven Pressfield, The War of Art: Break Through the Blocks & Win Your Inner Creative Battles

Another quote from Pressfield also attests to what it takes to “go pro”

“The artist committing himself to his calling has volunteered for hell, whether he knows it or not. He will be dining for the duration on a diet of isolation, rejection, self-doubt, despair, ridicule, contempt, and humiliation.”
Steven Pressfield, The War of Art: Break Through the Blocks & Win Your Inner Creative Battles

Last is a good synopsis of how a pro tackles his daily work. This is Pressfield’s schedule. Who is also divorced, by the way. And has no children to contend with. Yet he still works within disciplined time constraints.

1. W H A T I D O (Pressfield)
I get up, take a shower, have breakfast. I read the paper, brush my teeth. If I have phone calls to make, I make them. I’ve got my coffee now. I put on my lucky work boots and stitch up the lucky laces that my niece Meredith gave me. I head back to my office, crank up the computer. My lucky hooded sweatshirt is draped over the chair, On my thesaurus is my lucky cannon that my friend Bob Versandi gave me from Morro Castle, Cuba. I point it toward my chair, so it can fire inspiration into me. I say my prayer, which is the Invocation of the Muse from Homer’s Odyssey, translation by T.E. Lawrence, Lawrence of Arabia, that my deal mate Paul Rink gave me and which sits near my shelf with the cuff links that belonged to my father and my lucky acorn from the battlefield at Thermopylae. It’s about ten-thirty now. I sit down and plunge in. When I start making typos,
I know I’m getting tired. That’s four hours or so. I’ve hit the point of diminishing returns. I wrap for the day. Copy whatever I’ve done to disk and stash the disk in the glove compartment of my truck in case there’s a fire and I have to run for it. I power down. It’s three,three-thirty. The office is closed. How many pages have I produced? I don’t care. Are they any good? I don’t even think about it. All that matters is I’ve put in my time and hit it with all I’ve got. All that counts is that, for this day, for this session, I have overcome Resistance.

 So that's my advice.   There seems to be an undisciplined artistic drive present, yet an artistic sense  nonetheless.    So I've presented some stuff from Pressfield as food for motivational thought as well as how to approach things.       These are similar issues to mine.
You still need a break from your work.     So Listen to your conscience and step away from the table.   To establish distance from your project/work in order to regain proper perspective.    This is not about more or less detail.  This about over attachment and disordered priorities, at the moment.           Just my two cents since you posted. 

 Wish you the best.

It occurred to me that you might want to read these articles by the Russian designer Art Lebedev. He’s a designer and not an artist (at least on this site) and he’s one guy where you can be sure of the distinction. However, the process of creating art and creating a design can be similar. Many of the mental heuristics are useful in both spheres… so I would say, read some of these:

http://www.artlebedev.com/mandership/

Starting with this one, it directly addresses your problem:
http://www.artlebedev.com/mandership/167/
and
http://www.artlebedev.com/mandership/162/

And finally just remember that details are always in service of the whole. You’re not wrong to want to put in all that detail. It’s my belief that your level of detail needs to hold up down below what’s visible, otherwise the eye can perceive the missing detail, even subconciously… but convincing =/= accurate and good =/= convincing OR accurate, so keep that in mind. Don’t let them stop you from completing the whole.

Thanks for the support and ideas guys, I’m so glad I started this thread! It’s given me a lot to think about.

I think the root of my problem is I like to take my time with things that matter to me. If I build a 3D model of a real-world object I want it to be accurate, because I want to capture it indefinitely in a virtual world - that’s what it’s all about for me, the artistic side is something I admire but rarely strive for. As wrong as this sounds I get a lot of satisfaction from a photorealistic render, even if it wouldn’t be an interesting photograph.

Terms such as “speed modelling” fill me with dread. If I have to rush things to get them finished it takes all the fun out of the task, and it should be fun - that’s why I started it. Having said that, I will take what’s been said here on board and finish my Strat model. By “finish” I mean give it a neck and some strings, then think about the details. Part of me would prefer to make nothing at all than a bad model, which in turn makes me afraid to make mistakes. That’s why I haven’t posted a finished render here for… ever. Oh, but I do have a hard drive full of accurately-modelled bits of things!

I have the same problem. Finding the balance between what’s accurate and what’s necessary is difficult - especially without a deadline. 3pointEdit has the advice I try to enforce:

1.-Make a hard deadline.
2.-Stick to it.

Don’t think of it as compromise or inferior work, think of it as a trade off in accuracy for a gain in likelihood of completion. I’d like to say I do these things but inevitably I’ll end up tweaking a texture for an eternity while the rest of the scene remains empty…

Step away from the realistic imagery for a change. NPR stuff like toony, abstract, whatever, for quick and easy output. You already have the skills/reference to create many things in Blender that don’t require absolute realism and massive amounts of time. Have some fun and crank out some product that doesn’t require precision for a change. It will reset your brain and forge new habits.

-LP