[QUESTION] Art vs Coding

[Shinrael]

Once again, sorry if this is the wrong section for this type of a question.

My question is: Which takes more time? The art aspect of WoW or the coding ( programming/scripting/coding w/e you call it and w/e it is. )
And which is harder and requires more qualities ( skill, creativity, knowlegde, etc )?

It's because me and 3 classmates had an arguement on this topic. I got thorn apart because it was 3v1. Even if I said something legit they just ignored it and kept mocking me. I can bet they didn't even know/care what the truth actually was ( 2 of them ), they just wanted to mock someone.


Please include proofs in your comments. I really have to know this. Also, keep in mind that the 2 questions are separate. My personal opinion is that coding is harder ( requires more skill, knowledge and so on ) but ultimatelly creating new zones and tier sets takes more time no matter how good you are because the continents are huge and the items have many categories and classes to satisfy.

[Larynx]

Art probably requires more resources. Coding is just tedious, but it's just a lot of proof-reading and copy and pasting. Which takes more time? It depends on the length of the code and the model. Concept art? An hour of diddling. Large 3D rendering + texture? a lot more time. Same thing with code. The two things aren't exactly equivalent comparisons.

I don't believe there is an objective metric by which we are able to judge what is harder, especially without having defined what "hard" is.

[Golden Yak]

It'd be difficult to determine which one is individually more or less 'harder' - they are wildly different. They might be easier or more difficult depending on the type of person attempting them.

In terms of what takes longer, in my experience art is generally completed before the code. It can really depend on the game, however, and how much art resources it requires versus how simple the code base is.

In something like WoW however, given that we're talking expansions rather than a brand new game, art is likely taking up more resources - the code base is there and might need altering, but the basics were put in back in vanilla. New content requires a ton of visual elements but not necessarily a lot of new coding.

[Deleted]

Doesn't the art side require a certain amount of coding also?

[Hyve]

Totally dependent on the situation and needs.

If a whole new tool is needed for some new service, or new zone design, or a new engine upgrade, then coding is going to take far more time. If it's a new Dungeon, or Raid, it'll be Art.

- - - Updated - - -

Doesn't the art side require a certain amount of coding also?

Yes - What a lot of people don't understand is that there is a lot of Coding Knowledge needed if you want to be a Creative (Art, Music, Etc...) Developer in the Game Industry. It's not just about drawing an image on a napkin, there's a whole lot more to it then that!

[SodiumChloride]

Cost-wise, art. Games these days are so detailed that you would need an army of artist to create all the art assets.

As for code, depends on your programmers. Also kind of unpredictable IMHO, sometimes you get hard to track down bugs ...

But overall, art dominates cost-wise, which is why developers/publishers are so eager to go multi platform. While you probably need a whole new engine, you can reuse the art - maybe a few tweaks are required.

- - - Updated - - -

Doesn't the art side require a certain amount of coding also?

There are normally specialised people whose job is solely to bridge the gap between the art team and the technical teams.

[Hyve]

There are normally specialised people whose job is solely to bridge the gap between the art team and the technical teams.

Aardvark Swift, a leading Game Industry Recruitment Agency did a talk recently at my University saying that the gap between Art & Code is tiny now. Artists are expected to be able to do that gap themselves, and programmers are expected to be able to do 3D Modeling, Art & Animation.

There are no longer just Programmers & Artists on teams, unless they're really, really specialised (Software Tools, Concept Box Art, etc...).

[SodiumChloride]

Aardvark Swift, a leading Game Industry Recruitment Agency did a talk recently at my University saying that the gap between Art & Code is tiny now. Artists are expected to be able to do that gap themselves, and programmers are expected to be able to do 3D Modeling, Art & Animation.

There are no longer just Programmers & Artists on teams, unless they're really, really specialised (Software Tools, Concept Box Art, etc...).

Well, I don't know how that is going to work out.

Programmers normally can't draw/model for crap. Artists ... are artists. Creating 3D engines is not something you can just pick up over the weekend, it's very technical and very few people can do it well.

[Hyve]

Well, I don't know how that is going to work out.

Programmers normally can't draw/model for crap. Artists ... are artists. Creating 3D engines is not something you can just pick up over the weekend, it's very technical and very few people can do it well.

Aye, but that's how the industry is going. Programmers are not expected to produce Concept Art, but are expected to produce napkin drawings for designs, and 3D Modeling & Animation is a skill required by most Programming Positions in a lot of companies.

As for Art, they too need to be able to do programming.

Speaking with the people of Aardvark, they said the gap is small, but the specialisations are getting more complex. You're no longer a Artist, or Programmer, you're a Creative Developer who specialises in; Art / Modeling / Programming / Tools / Etc...

[Deleted]

Aardvark Swift, a leading Game Industry Recruitment Agency did a talk recently at my University saying that the gap between Art & Code is tiny now. Artists are expected to be able to do that gap themselves, and programmers are expected to be able to do 3D Modeling, Art & Animation.

There are no longer just Programmers & Artists on teams, unless they're really, really specialised (Software Tools, Concept Box Art, etc...).

Well its really not hard from the programmers side.
Takes a day to learn some basic modelling and if youre a programmer who does the engine you will know everything about the tech behind it which makes understanding easier.
Of course the results will look like shit but thats another issue and entirely insignificant when it comes to understanding the artists.


Regarding time art vs coding. It depends entirely on the task and the tools. A good basis is more important than the people. be it either art assets you can reuse or code/libraries that help you get shit done fast. I would however say that art dev time can unlike coding be cut significantly by skilled artists . Some artists do the same picture in 1hour where others need 4+ cause they overpaint and do all kind of tricks to get the same quality the first guy can archieve from the getgo. You can't do that with skill in programming, speedup here comes more from better tools, preexisting code or autogenerated code than beeing skilled in writing excellent algorithms. Infact often noone cares, even if my code is 50% faster, if i just decreased cpu usage from 20% to 10% who exactly will notice? noone.

[Pull My Finger]

It's impossible to compare, really. The disciplines are so different, and most of them highly complex nowadays. Typically, people also tend to grossly underestimate the skill set and time exposure involved in at least one of the disciplines. Frankly, I've already seen outrageous nonsense in this thread (concept art takes a hour of diddling, or: coding requires more skill and knowledge, for example)

One thing that has been already said is definitely true: the gap between art & code is a very tiny one today. I'm currently studying at a place where coding and art is very intertwined, and I can definitely confirm that. In today's game making, almost all people involved are expected to be very tech-savvy. You'd be surprised what full blown cracks the audio guys are, for example. Overall I can also say that serious, high quality art takes crazy amounts of time and talent alone doesn't get you far, and while almost anyone can learn to code and to make things work, there are gigantic differences in the quality, elegance, upgrade-ability, interface connection and ressource usage of code.

[Huehuecoyotl]

Art probably requires more resources. Coding is just tedious, but it's just a lot of proof-reading and copy and pasting. Which takes more time? It depends on the length of the code and the model. Concept art? An hour of diddling. Large 3D rendering + texture? a lot more time. Same thing with code. The two things aren't exactly equivalent comparisons.

I don't believe there is an objective metric by which we are able to judge what is harder, especially without having defined what "hard" is.

I can tell you have never coded, from the way you describe it. There is a lot more creativity to a good piece of coding than people who don't do it realise.

As to the question; it has no answer. It like asking if the night is darker than an elephant is heavy. I suspect the coding requires more man-hours than the art aspect, but I don't think this is the main thrust of the argument you were having. Both aspects require skill, creativity and knowledge. There is no metric to compare the levels of these for either aspect.

[SodiumChloride]

Aye, but that's how the industry is going. Programmers are not expected to produce Concept Art, but are expected to produce napkin drawings for designs, and 3D Modeling & Animation is a skill required by most Programming Positions in a lot of companies.

As for Art, they too need to be able to do programming.

Speaking with the people of Aardvark, they said the gap is small, but the specialisations are getting more complex. You're no longer a Artist, or Programmer, you're a Creative Developer who specialises in; Art / Modeling / Programming / Tools / Etc...

Tools would fall under Tech Art department no?

I have no idea how they can claim the gap is "small".

I doubt anything a programmer produces with a modelling program is anything you would want to include in your finished product. >.> Same for artist code. The programmers won't want people who don't know what they are doing writing any production code if they can help it.

"Creative Developer" sound like a buzzword to me ...

Look at Apple. Does Jonathan Ive do any electronics or software work? No. Neither does the people in their hardware or software group do any "industrial product design".

[Pull My Finger]

I doubt anything a programmer produces with a modelling program is anything you would want to include in your finished product. >.> Same for artist code. The programmers won't want people who don't know what they are doing writing any production code if they can help it.

Isn't that completely obvious? You can't get any shit done if the guy who's supposed to deliver one piece of a highly complex puzzle doesn't have a clue about how the big frame works and is unable to deliver his stuff in a way that it can be seamlessly implemented. Ever done any game or application development? It becomes very clear very fast.

[hrugner]

The question is wrong, which makes responses difficult.

Coming up with an elegant and useful programming solution takes years of work and a strong understanding of the systems being worked on. It doesn't just pop out, and isn't always obvious. Art can be the same way, you don't just get to sit down and make stuff up, you spend time researching and look at other solutions to the artistic problem facing you. Then there is art that's just programming, like visual effects design, fractals, response timing; where does that go?

Essentially, the conversation you're having implies a pretty amateur level of understanding of both tasks.

[Danner]

I cannot speak of artists at all, since I have no artistic skill whatsoever. I can speak quite a bit about programming, and what is required for getting the two together, as I have been working on this quite a lot. I don't work on any game engines though, nor in the game industry - but it's close enough.

A master programmer can replace 50 coders fresh out of university. That is the value of experience in the field of programming. Programming is a skill that takes 10-15 years to master (if lucky). I personally feel writing good code is about 30% ingenuity, 20% elbow grease, 20% artistic gutfeel and 30% theoretical knowledge. Mastering the ingenuity parts require experience and intelligence, to come up with a design and code that works well. Mastering the elbow grease parts is usually about pulling long coding sessions without slipping up in quality. Mastering the artistic gutfeel is about making designs and interfaces that can be implemented, and can stand the test of time. And the theoretical knowledge will heavily color everything you do, allowing you access to better and smarter solutions. I personally think writing good code is an artistic profession. No two code pieces look alike; and some pieces of code are undeniably beautiful to behold.

Being a programmer is a though profession though. Deadlines for finishing are usually unreasonable. Most programmers tend to burn out before they hit 30; advancing to designer roles or switching professions altogether. Unless you are lucky enough to find an employer that understands this burnout cycle, you will join those fates if you pick up this profession. But since this burnout cycle is quite a brain drain on the knowledge base of programmers, you'd be surprised how many programmers doesn't even know how to use a debugger! I personally find this totally unacceptable, but alas, I am not the one teaching the next generation of programmers. Best I can do is help the ones working at my workplace.

I like to think that a team of 3 master programmers can complete any unit of code. The problem with relying on 3 people is that when one of these quit or is hit by a truck, you are kinda screwed. So most people tend to have larger teams. From 10-15 coders, to even a few hundred in the extreme cases. But I like to think that this team meets the same deliveries as the team of 3 master programmers. It's just a slightly more robust and far more expensive team. =)

The wow game engine (ie, the code rendering the gameworld assets you see on the screen) for example, could easily be built and expanded on by such a team. But, you need an apparatus to support them. You don't want to bog them down with support - so you need a large support team to hide all these things, identify issues, and raise the important ones properly. You don't want to force the team of programmers to handle regression testing and quality assurace (though they will have to verify that the code they wrote works, you don't want to force them to spend 95% their time on it testing that their latest addition works on all the 2500 different GPUs out there). So you need a testing team (most companies drop QA/testing teams, for poor results). You need a proper management chain that can avoid panicking over every single issue (clue on how to not manage is "some forum post says that this and that mark of an Nvidia Card crashes under this and that circumstance. Panic NOW and drop everything you are dong, and fix this!"). And most importantly you need the right kind of people in the team. You need people who can cooperate, deliver, drive their own work, while still having ambition to make that engine better.

More important to know is that programmers merely enable possibilities for artists. Artists by nature want everything. They want to be able to use an infinite amount of textures per polygon. They want to be able to use an infinite amount of light sources per fragment. They want normalmaps, diffusemaps, lightmaps, specular maps, dynamic soft shadows, displacement maps, an infinite number of vertices, and gods knows what. Some of them get downright grumpy when you tell them that if you add all this to the engine, things will run at 0.2 FPS, and will just angrily reply that in that case the game will look crap! So the engine guys and the artist guys have to decide what is the most important, and work out some compromise. But as the programmer, you have to reach further to enable them a game engine that allows support of more features, to make that compromise as good as possible for the artist. That way the artists can make things look better. How you support it is the question. So the programmers ask themselves if they can make those soft shadows slightly faster? If they can make some sort of system that easily allows to use normal maps on only faces that actually need it. They think about doing stuff like LOD models to allow artists to use more vertices. But if you add LOD models, then what happens to model load times? Do we want to spend 2 minutes in a load screen when HS'ing to Ogrimmar? Can stuff like this be loaded in the background? Can it be compressed so that the loadtimes aren't as bad? There are tons of things to consider here, and always something that can be done better, smarter or faster. Which can always be translated into "looks better".

Then there are tools. Converting data from the artist tools (like 3D studio) to the game engine, and related assets like textures, skeletons/animations and whatnot. These tools usually do heavy model optimizations to increase performance, and they are a major amount of work to write and maintain. A small improvement in how data is used can easily turn into a small % FPS gain, which in turn means you can use more quality assets - but it also means you likely have to rewrite the tools. Or even making the tool faster or have better speed can easily save a man year in artist time in the long run. There is tons of research being done on graphics as well, so you want to read up on what the world of academia cooks up as well if you want to excel. In fact, you probably want someone devoting themselves to just catching up on research rather than coding if you want to make a visually spectacular game.

In short, the programmers will have a lot of work. And that's only in the game engine department. You'll find stuff like this all over the place; from database design, to user interfaces, to login security, to combat and ability mechanics.

--

And even with that wall of text... the artist work is just as hard. I just don't know much about their work process. But I totally respect someone able to make an orc look good using only 3500 polygons, single texturing and no lighting (or there about, I'm guesstimating here). That is a workfeat that can't be done overnight. That said... if you think the artist's job is done there... it's not.

Someone has to put all this together and verify that it's actually looking good as a whole (art direction). That's also an artist job. It may not be the same guys that made the orc model. It may even be one of the senior programmer guys doing this. But it's still a job for artists; someone with a sense of art.

[SodiumChloride]

Isn't that completely obvious? You can't get any shit done if the guy who's supposed to deliver one piece of a highly complex puzzle doesn't have a clue about how the big frame works and is unable to deliver his stuff in a way that it can be seamlessly implemented. Ever done any game or application development? It becomes very clear very fast.

But do you want your programmer wasting his time trying to explain to the artist why something can't be done/what the trade off is? Do you want the artist to waste time trying to explain to the programmer what he wants?

That's what the Tech Art department is for, to translate artist request into a language that programmers can understand and vice versa.

[gaz770]

My vote would go with coding, art is so easy to do these days...

[Cosmic Janitor]

Depends on what you want to do, really. Rewriting stuff in your engine that impacts alot of other things may take ages to the point of being simply out of reach for your project. Doing something in your secript engine to make an encounter? Depends on the encounter. Quest? Depends on the quest.

Same for art, while I doubt you will come across alot of art assets which are downright impossible, you are still limited in many ways (mostly because of tech though). The actual time spend on art also varies. Models? Depends on the size and complexity and how much the other guys will bitch about your stuff. Animations? Depends on the tech you are using and the bitching of other people. Textures? Depends mostly on your tools and your skill. Concept art is down to your skill alone mostly, though having proper programs certainly helps.

[Yig]

Do you often sit and stop working on your code and come back to it and purposefully not work on it for days or weeks at a time and then completely rework it based on a moment of inspiration or new direction? Does looking at your code too long make you make blatant mistakes that make no sense, do you ever look at your coding backwards or rotate it around, do you have other coders look at your code and suggest tweaks here and there in your code, maybe even trading that code off with each other and jamming on the same code with your own personal technique?

You don't just sit down, decide you need to draw X, then draw it. Sometimes you do, often it's a bad idea. And you don't just take anyone who can draw something better than someone who can't draw and give them a job. It's often sad how the smallest bit of artistic ability has people surrounding you telling you how famous you will be when you grow up just because you can give a drawing a personality. It's all magic to them, and apparently you can just flip a switch and it works.

And to the person who said "art is so easy nowadays", as opposed to when?

These two issues are really not very analogous and any number of arbitrary factors will radically alter the answer because context matters in the real world, this isn't an absolute game of rock, paper, scissors.