Unity Cannot Produce Good Graphics....I Think Not

The games industry should focus less on graphics and more on sound. HD Graphics are nice, but without comparable sound, the immersion that the graphics try to create is broken. Sound engines should be more akin to event based, realtime DAW's. If that needs hardware, make the hardware.

 

Im really not sure where this "unity games aren't pretty" rumor came from. Sure, you need to put a bit of work in (compared to out-of-the-box fidelity that many other engine offer) but the ability is still there. I also think that 'layering' of effects is unhealthy. By this I mean using multiple image effects, reflection engines, IBL tools and so on in one project. You might need them but its always better to look into one method of achieving all of the effects you want rather than multiple offering only one each.

 

None of the videos represent unity or would represent any games dev software particularly well, nothing that impressive

Do graphics matter? Of course they do, so long as they're at least functional, good, if they look spectacular, super, but you dont like anything in any other medium by just one factor so why suddenly now? The whole 'interactive stuff' evolution has been dragged down by these childish notions of of what a media can communicate. How are we still at the point where we're glad there's graphics, rather than them being part of a whole work of usually highly collaborative art

The can *insert name here* do great graphics is an irrelevance, and i think focus should be put more on the opportunities the medium gives us on all sorts of levels, take a few games like say, Journey or Ico and these are just beautifully designed game that happen to be beautiful, thats where you decide what good graphics actually are, and thats interesting to ask people as 'good' should not mean 'realistic' and wether a realistic or stylised approach uses more tricks or more horsepower is irrelevant

I think personally, id like games to lose robots, cars and guns much more often (Theres irony given my current ambition there), id like people to get over realistic graphics as soon as possible, its incredibly boring, as a friend once said 'i think they should stop making games with humans in them' and increasingly I think he's right, this need to replicate the real world in some horribly contrived narratively abysmal exercise in shader technology. Ive played the last batch of sequels merely to try learn something and there was sadly nothing to learn, in fact they showed still a tendancy for smoke and mirrors tactics that i wouldnt regard as 'good graphics' at all

Unity can easily, without argument, deliver as much and more of what has been the visual output of the most critically celebrated games of recent years. It can do more than that, it can blow you away, and give you a ton of ways in which to do that from the very start. If you're looking for the 'good graphics' button in unity, your art is going to be terrible, and it will be your fault. If you approach your visuals with the competency, intelligence, imagination, technical and traditional approach of the finest games you'll play on any platform, unity can probably do it, and no gi in sight.

The big thing with these arguments is how little understanding of how much it even makes sense to do something as asinine as poking at Unity's graphics? I would hope these people do not play my games. I know how to make pretty things (much better than im able to make a game that isnt just rubbish to play) but i was trying a concept, not 'good graphics'. Maybe the kids should go off and play with whatever their favored flavour of the month is regarding making games and try actually make something, and then they can realise they're not really up to it and theyre making an embarasment of themselves in public and they can go and study to be an accountant or work in macdonalds or something

 

Umm. With my eyes.

 

Ha you know what I meant ..

Getting a decent game done is what's most important and for that you don't need stunning environments. But if were talking raw cutting edge graphical power, then Unity isn't at the forefront of it. Whether people care or not is a different matter, it's not as if Unity graphics look like Final Fantasy 7.. Man that game didn't age well ..

Optimisation / asset streaming and 64-Bit editor's a different matter..

My favourite game to date is Dragon Age origins built in Eclipse engine and that's originally 2006 tech? Still looks good enough today and the gameplay is excellent.

 

Unity will probably always be able to be used in a way that evokes an effect another engine might be all about with probably a bunch less efficiency and accuracy but the only people who care in the slghtest are the people making the game afraid of what their peers making other games might think of them. Thats the only reason i can think of this argument to continue while being so fecking dumb. Generally what you can do with a system (lets call it high tech, an integrated gi system) can be approximated with spot effects (lets call that basic, just putting lights there instead) and noone would be able to tell the difference or if they could, care, and it only being in a ridiculously narrow spectrum of the possibilities that people ever scrutinise so much (lets call it fps, motorsport and rpg)

Theres a fellow who works for Pixar (was the lighting lead on the blue umbrella, if youve seen monsters university) who did a great talk on a deviantart chat session about how he created scenes that he was using for videogames. Very weirdly, a few years later, he was hanging out a bunch in my chatroom and i think its simply because of a certain girl who was a reg but hey, i got to quiz him about a bunch of things over a few weeks and he essentially echoed above sentiments. He doesnt care for indirect illumination, does everything with direct lights, in personal pieces i guess, the blue umbrella is probably the most realistic cgi ive ever seen until ive seen gravity so im guessing all sorts of fancy gi things were going on there, the idea is to use whatever it takes to get the intended effect

Pity 'Uses well judged considered visual motifs to absorb you into a spellbinding story of love and revenge' doesnt go down with folk as well as 'DIRECTX11 EXPLOSHUNS!!!!'

 

I agree. I don't consider myself a graphics snob. In fact, I love the trend of reduced graphics load in exchange for deep game play or creative game play. But there are good games out there with mediocre graphics that I can't get in to because of the presentation.

 

I think DA:O is a good example of a game which is both technically compentent and yet as pretty and charming as a worn sock. Which is a pity since concept art for the game looked great. I'm one of those weirdos who liked DA2 more anyway. 8)

Anyhow, I would say graphics are super important but high fidelity isn't. Thomas Was Alone is one of the best looking Unity games. But high fidelity obviously sells.

 

True, so many are trying to make good graphics. Try making retro graphics!!! lol

 

I like the obvious pixel art support on websites and that people still do that stuff but its funny how everyone i know who does pixel art wasnt born when pretty much the only cg art was pixel art

 

Im tired of pixelart now...i really ma, and im sorry to say it! i used to absolutely love it, but it has become a trend just like zombies....now all we are missing is a zombie pixelart survival openworld online game, i think theres like 10 on steam now at the same time, someone make it, u will be rich!

seriously tho, i couldnt care less about graphics, im not impressed at all by cinema looking games anymore, because the material they have to sacrifice in order to get it like story, gameplay...it seems to dissapear completely and u feel like u have done it before.

i play starbound now, it rocks, its pixelart but its better than any other game,why? because im probably gonna log 1000 hours in it...probably more!

it has SERIOUSLY good gameplay and progression, havent played a AAA title that even comes close...

 

The engines do not produce good graphics, the artists do!

 

Alright, so if I give you ID tech one your saying you could create just one simple scene that matches the likes of CryEngine?

Shader tech, rendering paths, shadows, SSAO, DOF, Bokeh, HDR, Bloom, Vignetting, color spaces / color grading, tessellation, parallax occlusion mapping, LPV, IRV, RLR, GI, Eye adaption and Anti-aliasing has no effect on graphics.

Well who knew? I do agree that a huge portion of the equation is the artist.

 

Can a paintbrush create a painting for me?

 

It's not a valid example. Because the engine has to match the hardware of the time. In all common sense, it is clear and obvious we must compare present day hardware from DX9 and up.

It's clear as day Unity lacks a couple of things, but there is no gulf here great artwork cannot confuse the picture considerably. Any layperson would struggle to tell the difference between a Unity title with marmoset shaders and a Cryengine game. While there are limitations, it's nothing like what people are suggesting.

All things equal, why do you think that a large majority of unique UDK projects, XNA projects look like crap? Because the art is crap.

The reason you don't see any cryengine indie crap is because crytek put a stop to it. And If you just dump user media in and you refrain from using Crytek's fabulous art, you end up with a total turd. A very shiny, polished turd.

 

The problem is right there at the thread title - PEOPLE ARE EXPECTING THE ENGINE TO PRODUCE GOOD GRAPHICS FOR THEM.

This is like asking the paintbrush to create the painting for the artist.

And then the argument goes into CAMERA vs ARTIST WITH A PAINTBRUSH - at the click of the button the camera can create the most realistic image with minimal fuss, but if everyone start using the camera, the value of "realistic imagery" becomes devalued - just like nowadays nobody will marvel at how realistic an image is if you take it with a camera.

It takes a real artist to create art.

 

No, it's more like people are expecting the tools to make the production of these things easier. Let's expand your metaphor a bit.

Using Unreal is like using a set of paintbrushes and a nice set of paints. You have a canvas, an easel, some primer, oil paints, watercolours, even some acrylics! By default, everything you make will have a pretty nice painted aesthetic, and the more time you've put into your craft, the better your art will get and the more efficient you'll be at producing it. When you start out, everything will look a little simplistic, but still good but as you go on, you'll start to be able to produce some far better looking results.

Using Unity is like having a box of pencil crayons, a sheet of A5 paper, and an art store next door. When you start out, things are going to be pretty rough.. As a starting point, you're not really going to have the same level of freedom that a painting kit with a full set of paints and brushes will give you. However, if you keep working and maybe visit that art store for some extra supplies, you'll be able to produce some considerably better results.

 

I absolutely agree and I like your approach to describe this matter.

Should be a sticky. But I doubt UT likes their engine described as default crayon box

 

I really liked that Murgilod, cheers! The asset store is unity's seemingly hidden game changing superweapon, its a bit absurd to see it getting so little coverage when dorks compare things whe its been quite at the heart of unity for me. Unity is just a lot more modular (and offers more direct specialising) than other software!

Thats a simplistic way of putting it, but yeah, if it wasnt for the asset store i probably couldnt be arsed to make games, full stop, i dont see any direct competition other engines are giving? Theres stuff you can do in a flash, thanks to the asset store that mean months of work in something else.

 

The asset store is absolutely fantastic, especially now that there's proper support for things like mechanim assets and upcoming stuff like Shader Forge. Shader Forge is a huge benefit for this whole graphics discussion because it's bringing UDK-like shader editing to Unity in a far more complete and updated form than Strumpy Shader Editor.

 

Well said, HippoCoder, I am da bawss,


How do you tell majority of the artists on the forum looking for jobs that their work is sub-standard and their artwork is crap, they bring poor artwork into the project and fans gets disappointed on the graphics expectations?


It's like you order fast computers, SSD-drives, AutoDesk Maya 2013, AutoDesk 3D-Max 2013 and Adobe cloud subscription licenses and the artists you interview, seems to give only turd, stick drawings and simplistic artwork.


How do you tell them to practice their art on their own private time, instead of putting other people's artwork into their portfolio, bait switch and other bad practices?


I don't know how to tell them that in a polite manner.


This whole Unity vs UDK vs Crytek about finding skilled artists able to competently make good art.

What should an artist know so they can create good artwork that fits well into the gaming environment?

 

Does or should (?) the artist know about SSAO, DOF, Bokeh, HDR, Bloom, colour spaces, tessellation, parallax occlusion mapping, LPV, IRV, RLR, GL and other other advanced shading features?


I asked around; the artists say that it's the duty of the developer. The developer says he's not responsible for the artwork and that's the responsibility of the artist.

 

You say the truth

 

Since i do both sides and have delegated in the past (this basically meant i was directing the whole thing when i shouldnt have been simply because i was the way the programmer and artists could communicate) ive felt that a competent artist should be fully aware of how shaders work, how to write/edit shaders and really have a decent understanding of everything under their remit. I would call that a competent artist, in games, and id consider a competent lead or programming lead to take responsibility for communicating to the artists everything they need to know so they should be able to figure something out for themselves than wait for someone theyve arbitrarily decided is responsible to fix it

Graphics is a particular field in which neither artists or programers should deny responsibility for anything within their remit

 

I'm wondering, how many of you actually bought the Pro edition ( And not used a crack to get it to work ) .
Even if i was satisfied with Unity3D graphics, I dont realy want to put 1500$ into this hobby ( Especially when the competitor gives better graphics for 99$ ) . I still like to work on unity, but when ever i do somthing i cant use normal shadows ( This directional light hard shadows looks like crap ) or simple glow effect, This is frustrating...

 

@ sahkan

I don't really understand the attitude, your complaint is regards to the pricing not suiting you when both unity pro and udk are priced with your licensing in mind. If udk does all the stuff you want and its only ever going to be a hobby.. why not just use UDK then? People paying for pro know why they are paying and what they are getting, and thats a very generous price compared to what a udk license will cost you after your game is released, assuming your game makes enough money to support a 3 person team full time

Nothings binding you to unity, you chose the most appropriate tool for the job, you get unity pro with an idea to making money with it right, so continued use of free is an investment to learn things that make pro a good monetary investment

 

I guess this is because they have smaller financial back than their competitors, but this is not somthing the customer should care about .

 

The basics - scale, mass, proportion, composition, balance, space, style, lighting, contrast, emphasis, focal point, motion, rhythm, continuity, depth, shape/form, texture, color, symmetry, asymmetry, duality, unity....etc etc.

As for the technical side - efficient use of polygons, textures and materials/shaders, and whatever the tools the artist is required to use to complete his job - eg. every feature in Unity - SSAO, DOF, Bokeh, HDR, Bloom...etc etc. The technicals is only to compliment the fundamentals. You can't be an artist to know only the technicals - or else you will just end up making "programmer art" like below.

A lot. There are enough of us who bought Unity (and the add-ons) that Unity can afford to keep over 200 people employed around the world for years (until the next cycle) and still turn a profit.

It's the entitlement generation mentality ....... "I want what I want, but I don't want to work to get it".

 

Come on now, if you are caring so much about people who needs money there are always people who needs it more than Unity's employees you can be sure of it .

Actually unity will give you lisence to make money but no more than 100,000$ in the free version .
UDK will only begin to take 25% Commission from the money you make after you pass 50,000$, so the different is 50K compare to 100K, when you reach that much, i dont think that this is too much of a problem .

Personally, I think you should learn both, and chose the engine that Obstacle fit your project better, cus if you want a small mobile or net game, the graphics dosen't matter that much, but if you go with tripple A quality, you go for UDK, that way you dont have to buy the Pro version for better graphics, the only Obstacle here is your amount of knowledge .

 

Well at least I'm glad that statement picked up some movement, because in a nutshell IT ALL MATTERS. Every part of the process is as important as the other, terrain textures suck there's and issue, you over do the SSAO it makes everything look crap, your character models look poor it's an issue..

There all little step's that add up to a final result, the artwork, level design, animation, coding, UI, Post processing effects..

Maybe because I dabble heavily on both sides, you realize that after extensive polish from engine PP it just doesn't look like the same game anymore. There is no excuse for poor art, but let's pretend for a sec artwork is of AAA quality and we switch them between engines. Can you tell? You can't half tell a difference, when I did a test with Micheal O's asset's in CryEngine they didn't even look like the same assets.

Then again, I have seen some CryEngine games that have fantastic beautiful terrain spoilt by extremely poor character models.. On the flip-side I have seen some beautiful games made in CryEngine by Indie's I mean truly stunning which I have never seen in Unity. Whether they actually ever got it released, the odd's are probably stacked against them but that's a different debate.

So where's the deficit? How come one guy in CE is making the likes of this, by himself whilst there's tons of other great examples. Yet I've still to see teams with potentially much better artists do the same? (All rights to Halcyon)

 

Of course not, why should you care if they starve and can't feed their family or not right? You should always get stuff for free right?

Unity is priced as reasonably as it can be. If you can't afford Pro version upfront, you can always pay $75 for Unity on monthly basis. $75 a month is really NOT MUCH. A night out drinking with friends will usually cost that much at minimum over here. Even if you live in some impoverish third world country like Nigeria - where annual wage is $1377 USD - which works out to be $114 per month - even they can afford Unity (assuming one of the partner/spouse is also an equal income earner...)! If not, you can always save up. Takes maybe 2 years if you are in Nigeria.

 

Or pirate it and buy when you want to release cause its easy enough for unity to check if the license is valid, i think ive been using the internet too long a time to sustain any bullshit about piracy, and unity's own mechanism protects it. You buy pro to make money, you can only make money if you buy it, simples

 

I use Blender, Gimp and such...
I'm trying to find the best combo of starting a project with with smaller Expenses as possible .

 

So, I am curious, where do you live, do you have a job?

 

Our artists know about color spaces as they have to as they work in Linear space, they also understand what tessellation, Global illumination and shaders do. Because you have to understand it for correct modelling (or at least do it and not understand it), let's say you open 3DSMax and you apply a material, within that option you have to select what shader scheme for that material, which for CryEngine you have to change them all to CryShader (Just an e.g.).

As for the rest they understand relatively what it does and the impact, as we sit with the artist, see how a model / modular pack looks in the game and if any tweaking is required. We are only a team of 5, but AFAIK that's how Bestheda do it as well.

 

And, if you're a student, you can get it half price. Perpetually. Or you could always just make a decent game with the free version, release it, and get the $1500 that way.

 

It's fine talking about the licensing scheme, but have you ever tried to use UDK to complete a full game? It is very difficult compared to Unity, so you get some better shaders / shadows and lighting to name a few. It'll mean little when you tear your hair out, Kismet will only get you so far dependent on what you want to do.

Then it's time to dive into Unrealscript, what I remember of UDK it doesn't strictly support a Biped system and trying to get a main character into your game and sort out the animations with UnrealScript can be painful to say the least. It's easier to just to make your own shaders in Unity, with Unity you can kind of glide across not really knowing what your doing, in CryEngine and Unreal you don't have that option. So for beginners the learning curve is steep..

More power to you if you do it though, you will learn a lot very quickly.

If you ever get to a point, where you care about the licensing then you're in a good position. As much as any one of us can or will complain about it, Unity does something amazing. It allows small teams to do a lot very quickly!.. Now that's worth more than the $1500.00 price tag IMO.

 

Israel, I do have a partial job and i study ( unity, udk, blender and such ) in my days off, i rather find ways to save my money instead of use what i have .

I know, i'v tried to do it all with kismet and failed, I'm in that part of playing animation on the character and almost went crazy ( Why the hell cant i just someSkeletalMesh.play(w/e) ?! ) . they have their own ( complicated ) ways to do that, was just reading this : http://www.moug-portfolio.info/udk-animation-basics/ while posting in this forum .

I'm sure once you are an expert in it, and learn why their way has their own benefits, you wouldn't think twise on what engine to use when ever you want to make a big triple A quality game .

 

Some of the job candidates we get here are on 40+ weeks of welfare and they are 24 - 25 years old.


They don't want to take an intern job to learn the skills of the trade nor take their own private time to learn and become proficient using Unity free version.

What other things should a game artist know? For instance, should they know about light-mapping, what level of proficiency - photoshop expert, Maya or 3D Max expert?

 

Light propagation volume (Indirect lighting), Irradiance Volume (How do I best describe it? It's like mixing light's to create a simulated radiosity effect which also adds color to reflected light) , Real time local reflections and I think GL was supposed to be Global Illumination .

Lead graphics artists with next gen in mind would probably need to know how to manipulate things like irradiance maps (Vray) and make the best out of all technologies, although it's going to be the techies that implement it in an actual engine and level designers / devs who use it.

For screen art it may be the artist who does it all, just depends on what your doing / trying to achieve.

 

That entirely depends on the position and pay scale
Most of the 3D artists nowadays knows about most of the jargons of CG graphics like SSAO, DoF, Bokeh, HDR, Bloom, colour spaces, tessellation, parallax occlusion mapping, because these features are now standard on most AAA games and graphics card manufacturers are touting these features for years. Although I must say I don't know what LPV, IRV, RLR, GL are, to be honest... (and a quick google doesn't show up anything related to CG either so I am guessing its not in widespread general use)

But to be honest, I don't think knowing those advanced shading jargon matters, it is a good thing to know - but not essential (not unless you are Lead Technical Artist) - a good artists you look for are the ones who have very strong traditional skills and understand the fundamentals I mentioned before - so when they show you their portfolio - like their ZBrush/Mudbox/3DSMAX/XSI/MAYA/Blender/Whatever models, it will show. Strong sculpting/modeling/texturing skill (or animation) with attention to details - and that is ALL there is to it. All those HDR, Bloom, SSAO, DoF are fluffs that don't really matter when you are judging a candidate. And that is essentially what you look for in an artist in this industry.

 

Ah, I know these terms, I just didn't know the acronyms of these. You Y generations and your acrynoms and twitters. Tsk tsk tsk

 

HA, apologies.. I do think with the next gen more artists are becoming technical, Vray has a lot of the mentioned next "Gen" lighting tech (Even though it's been around for years)..

Plus it's a great way to make your concept art look flashy. Irrelevant or not if the game developed is a bugged out sloppy boring game!..

Anyway I think we went a little off topic here slightly.

Q Can Unity produce good graphics?

A: Yes of course it can, get back to work .

 

Well, i'm not in the position to say what an artist is and does get judged on, im yet to apply for a position, but at any point in the artist vacancies at a studio i'd be disappointed if i wasnt expected to show an aptitude, an understanding or a clear ability to learn any aspect of the art part of videogame creation from sculpting to shaders and even at least an awareness of what s going on in a rendering pipeline

Any post effect shader i'd consider, personaly, a bad mark against you cause its fluff to distract from faults in ones work, i could only really understand hdr/tonemapping and colour correction as post effects you should have a good understanding of. Not sure i agree with shaders in a unity 'material' sense though, i think an artist knowing whats going on in all the shaders they'll be using would be quite a strong asset as they could easily add their own improvements to them under the lens an artist sees things (I'd feel happier a really good concept artist was governing the shading of something over than an art tech guy, having been the art tech guy, overstating my value, and the concept artist being an incredibly talented individual)

 

I'm referring to pay positions with all benefits and take home pay of $4,000 or more after deducting US taxes.

I'm not going to repeat myself so I'll post a link where you can read it.
http://forum.unity3d.com/threads/21...ase-developer-blames-Unity-for-failure/page10

 

Lazygunn, if you have ever worked with 3dsMax you will use shaders within 3DSmax.. This isn't anything to do with Unity at this point, also you will render with GI and GL in all sort's of modelling programs.. Now I'm not an expert compared to the others in the team, but I know how to make 3d meshes and character / models / rig them and print renders out for concept. Especially in other engines like CryEngine there is a strict procedure on how you create art for the engine and applying shaders is one them..

You don't need to really know tonemapping, work in SRGB and set the gamma and LUT. As long it's linear space when we receive it..

All that being said, you don't have to understand how shaders work as an artist.. You just need to know how to use them..

http://www.neilblevins.com/cg_education/linear_workflow_vray_max/linear_workflow_vray_max.htm

http://docs.autodesk.com/3DSMAX/16/...-A348-D0B72C472FDB.htm,topicNumber=d30e491947

 

Well, I feel sorry for you. And that's the problem when you are working with people OVER THE INTERNET where they can hide under false identities and proxies. That's the risk you take. I would recommend you hire LOCALLY - people you can actually meet IN PERSON. There is this layer of ACCOUNTABILITY when you are meeting people in person - of course that's not a guarantee people are not going to scam you or lie - but meeting face to face physically means you will at least know he or she can be held accountable to your local law.

Now, $4,000 doesn't sound very much - I assume that's a 1~3 months (depending on talent) contract job - I think normally a game artist on average make $40,000 ~$70,000 a year - $71,354 per year in the United States, according to the 2011 salary survey by Game Developer Magazine. Salaries for artists generally increase with experience and job level. Salaries for those with less than three years of work averaged $44,643 per year. Artists at this level require supervision and training by more experienced employees, and may only handle smaller parts of a game. After three to six years, they averaged $61,667 per year if they remained artists.

And like I said before - I would look at the portfolio of the candidate to see if he or she has strong traditional skills or qualities that's an asset to your team. And if he or she has the willingness to learn and adapt to your working environment and pipeline. Not knowing the latest jargon or tech is not a deal breaker - so long as there is a needs to be and he or she can learn and adapt. Also setup a test for the candidates (if its modeling, set up a simple test to see if he or she can model something of your choosing in under 1 hour) in the job interview. This will at least ensure authenticity of their skills. Also - most studios runs a "probation period" type of system so even after you accept a candidate - he will still only be working for a week (or a month) under real studio environment to test out if he or she can cut it.

 

Note: big graphics

@I am da bawss, these were the newer environments made. One of the artists here made these environments.

Do you think these are made using Unity or UDK or Crytek?

(Some graphics mess-up - they are not showing - you'll have to right-click to see the full-sized image)

 

If I am to take a guess, it looks like it. That projector shadow looks kinda Unity like
But its not a guarantee - it could be Irrlicht Engine. Or maybe even Quake 3!

EDIT: I see you change it to multi-choice question - Unity or UDK or Crytek... I would say Unity.

 

It's okay. We had an excellent black friday for our non-gaming division. Going forward, thanks for the infos. what an artist should know.