10 Reasons Why Your Render Sucks

#1

Hey guys,

I’ve been a member of these forums for several years now and I regularly check out the 3d stills gallery for inspiration. I recently realised that a lot of artwork is suffering from some very basic mistakes that are stopping it from being excellent. So I decided to address these mistakes in article titled:

10 Reasons Why Your Render Sucks

I’m sure I’m only scraping the surface here, so I thought I’d ask the community, what are the key problems you see time and time again in 3d art?

Here we go…

1. There’s no point It doesn’t tell a story, it’s not advertising anything and it’s certainly not pretty. What exactly is the viewer getting from this? Nothing makes me close the window faster than a piece of art with no clear objective.

2. You use pre-made content Stock models are great for studios who want to save time and money by purchasing a pre-made model. But it has absolutely no place in your portfolio. Personally I find no pride in showing someone a render that I haven’t created entirely by myself, but that’s just me. If you don’t know how to model it, why not learn?

3. You’re copying something far more successful I love Wall-E as much as the next guy, but that doesn’t mean I try to mimic what a professional studio has slaved over for years on end. Why? Because unless it’s an uncanny comparison (which it won’t be) viewers will only spot it’s flaws.

4. You didn’t plan it on paper first It’s easy to tell when an artist failed to put their idea on paper first: it’s a confusing mess. They started with an idea, skipped the planning stage and jumped straight to their 3d program. Most artists cannot model/texture/render in 3d at the same speed as their imagination. The best thing you can do is put it on paper as soon as the idea strikes you, that way you have a reference in 2 weeks time when you’re sitting at your computer and asking, “what was I making again?”.

5. It’s cliche If I see another cave troll or big breasted warrior I’m going to puke. Be original and create something that everyone hasn’t already seen a thousand times.

6. It’s a test render Hey cool, you just got your head around the new array modifier! Don’t post it on the net. Test renders are exactly that. Tests. They are a learning experience that should remain on your hard drive.

7. It’s poorly lit Let me say this once and for all: Dark is not moody. If you want to create a moody atmosphere there are plenty of ways of doing it, but making your scene dimly lit is not one of them. Pick up a copy of Jeremy Birn’s Digital Lighting and Rendering to learn how to light your scene like a pro.

8. You don’t realise it sucks No one likes receiving bad feedback on their artwork, especially after you’ve spent weeks creating it, but to tell the hundreds of posters that they “just don’t understand it” is like throwing salt on the wound. If you want to progress as an artist you need to be able to take critiques on-board and learn from your mistakes.

9. It’s boring architecture Archiviz is great skill to have under your belt. There’s a lot of work available and it pays quite well, but that doesn’t mean it has to be boring and emotionless. Read my post on 10 Architectural renders that break the mold or watch Alex Roman’s amazing short The Third and the Seventh and you’ll pick up dozens of ways to make still architecture interesting.

10. It’s overly post-processed There’s nothing wrong with fixing the colour levels or altering the contrast in Photoshop, but when you start adding filters and chromatic aberration to hide your own incompetencies there’s a problem.

Well that’s me finished, what are your pet peeves?

EDIT: Does this mean I think my renders are perfect? Absolutely not. My portfolio is riddled with flaws. In fact almost every single piece of my artwork violates at least one of these ‘rules’. This thread is about sharing what you’ve learnt in your journey as an artist.

#2

I think I love you.

#3

Cool post Redbyte. I don’t think I can argue with any of it. :cool:

#4

Ten reasons your renders suck:

#1: You are not using a linear workflow (gamma correction)

#2: You are not using a linear workflow (gamma correction)

#3: You are not using a linear workflow (gamma correction)

#4: You are not using a linear workflow (gamma correction)

#5: You are not using a linear workflow (gamma correction)

#6: You are not using physically accurate shaders

#7: You are not using physically accurate lights

#8: You are not using indirect illumination, but a bunch of oldschool “cheats”

#9: You are rendering to neutered dynamic range

#10: You are not using mental ray :wink:

(Sorry I was contractually obliged to throw in that last one :bounce: )

Seriously, though, the #1 reason peoples renders DO suck is that they are not understanding a linear workflow, and are fighting the computer at every step trying to compensate for this. It’s the alpha and the omega of rendering anything.

/Z

#5

:applause:

Agreed. Here’s one-

A lack of color theory knowledge. Some images’ choice of colors are nothing short of jarring to the eyes.

And thank you for bringing up the test render point. I still don’t understand this. If anything it should go in the W.I.P. forum as I could understand someone wanting feedback on how to make it look better. But it doesn’t belong in the gallery.

Yes, the glass cup looks amazing…but why do I care?

Ok, that was my griping for the day. From here on out I’m more positive :slight_smile:

#6

I have to agree with all of the above, but I wanted to clarify that last one. Poorly overly post-processed renders are annoying, but that’s not to say that 3d should necessarily do all the work. If you look at Third and Seventh’s behind the scenes, a great deal of work goes into color correction and other lighting adjustments. After effects is an amazing and often overlooked tool. You can save yourself a lot of render time if you know how to use it right. But again, I agree with what you are saying. People seem to find a few “tricks” like using unnecessarily strong vignetting and it really takes away from the final image.

#7

The terms linear workflow and pipeline make me want to kill people whenever I hear them.

#8

uh…uh…what are you trying to say?

lol…
Good points one and all!

#9

I agree with all that has been written, but I ask masterZap about you’re not using Indirect Illunation but a bunch of old school cheat, can you give some reference :wink:
Great thread indeed.
Jojo

#10

There should be a sticky of this in the gallery section. Very well done.

In terms of architecture, I would also say the same thing of car renders. There are some stunningly accurate and well lit car renders out there, but i’d prefer to see them in an interesting situation or in a story-driven action shot than just parked on some cobblestones in front of a cement building at midday. :slight_smile:

#11

This could be the best topic of 2010.

#12

Get in line…

I think you’re missing the point a little - an immaculately rendered image of a turd, is still a turd.

#13

Hi redbyte! great Post!

“Breasted warrior” jajaja! awesome critic!

You should post this under the title " 3D Ten Commandments!" :smiley:

#14

I agree with some parts, for example posting a test render of a shiny cube. But I guess every one goes this stage.

Anyway some parts sound a little elitist. The reason why a community such as this exists is to share knowledge, even if some people think it isn’t.

If some one posts a render of a huge troll with big breasts, a shiny sword and under lit background saying it’s ‘moody’ I wouldn’t just ignore him and close the render, I would give him some constructive criticism to help him or her how to better understand the image, to tell them why certain aspects don’t work and perhaps some of the parts that do work. If I didn’t, or if no one did that person wouldn’t improve.

This is the same for all of us, if we don’t receive external criticism and input about our work, regardless of what line of work that is, we will never improve to our full potential.

So, in retrospect; Please don’t post test renders of cubes or poser models, but if you have an image which you have put effort into and have tried to your best ability then please do post and hopefully some artist with more experience and a better understanding of lighting, colour and composition may just come by and help you on your way to making better artwork.

#15

#1 We use linear workflow. This has only recently gotten easy to do. “implemented in host programs” IMHO this still has a way to go though for ease of use.

#2 The mia_material or architectural shader as it is commonly called is a great shader. if you are doing “in-render” rendering. if you are splitting things up " as most small studios have to do" to be composited later then this shader is a nightmare to use and requires extra nodes and a LOT of extra setup at least in Maya and Softimage anyway. It is just not real practical to use. This shader has a very unfinished feel to it.

#7 yes but not always practical. this depends on what you are trying to accomplish.

#8 The fact is that most studios cannot afford the render-times that it takes to get the flicker out of animated GI. GI on a deforming object without flicker is next to impossible with MR without having 1 hour+ per pass render times at 1080p or higher. MR desperately needs so sort of adaptive cache like what is in FinalRender.

#9 Since we made full-float exr our standard output format it has made our lives alot simpler.

#10 Agreed actually except in the areas I have stated above it is really about the best thing going for the price. Especially on our chosen platform of Windows/Softimage.

But Love your works zap I have learned soooo much from you over the years. "even as much as it irritates me having to translate everything you teach from max to Softimage. I was really looking forward to taking you FXPHD class this semester but I just don’t have the time do to work. I hope you will offer it again.

#16

You should add (or maybe could fit in one of your top 10) : Learn to pose a character, learn how to make great silhoutte!. Amazing zbrush character often suffer from that.

#17

#11. Go to Eleven - Many images, and I’m very guilty of this for sure, don’t go that extra mile to put on the polish. The extra couple nights working on it can be the difference between a good image and a great one.

“5. It’s cliche” - is probably the biggest offense that bothers me the most. I can accept a terrible looking image, so long as they tried to do something interesting / unique.

By the way, this is a great thread! I will definitely be using it for my future projects to make sure I don’t fall into the lazy trap of making poor images.

#18

One of my pet peeves is when a guy posts a list of his top ten pet peeves but does not have the standard of work to back them up. Looking at your portfolio you are as guilty as the people you target.

#19

to expand on some of those points though.

You get a fair number of people who don’t have a background in ar, who do copy without understanding. What I mean by this is to say, you have an artist who is “inspired” by something another artist is doing, but does not understand the foundations of where the original artist was “inspired” from.

or to use a quote from Michael Crichton’s character Malcom in Jurrasic Park
“…it didn’t require any discipline to attain it. You read what others had done and you took the next step. You didn’t earn the knowledge for yourselves, …”

what you get is a lot of people standing on the shoulders of the accomplishments of what others have done without the drive, want or skill to go out and learn how to do those things for themselves.

That said. There are some cavets, I certianly don’t expect an animator to go out and become a super modeler or rigger. But having some understanding and knowledge of those skills will always be helpful in the end.

In the end the biggest issue I have is when people just don’t try.

#20

hahaha

not everyone can do briliant stuff,
all we learning to last days… and some of things like talent just cant be learned, but still is possible to do great stuff if working hard.

by the way - your portfolio arch.viz. fits to subject - u know that right ?