Attracting and keeping artists on an OSS game project

Discuss the development of other free/open-source games, as well as other games in general.

Moderators: Forum Moderators, Developers

VS
Translator
Posts: 187
Joined: November 27th, 2005, 10:07 am

Re: Attracting and keeping artists on an OSS game project

Post by VS » October 9th, 2009, 9:45 pm

Nobody has mentioned yet the possibility to render 3d models into sprites. This would have little use in Wesnoth which focuses on animations, but for games that need rather "just" rotated items - like transport simulators - it's a blessing. Model once, no animation, press "render", you're done.

I have been around Simutrans for some 6 years and during that time saw that at least 40% of all graphics for it posted anywhere on the internet were made by just 3 people (~2% of contributor population) who knew their Blender-fu. And all three of them are reasonably good with 2d as well.

The numbers alone should be enough to give some insight :wink:

ceninan
Posts: 4
Joined: July 15th, 2009, 3:56 pm

Re: Attracting and keeping artists on an OSS game project

Post by ceninan » October 11th, 2009, 12:12 am

It works quite well for animations too, especially when you need them in several directions. I think that eg. the Diablo and Baldurs Gate games used this approach, and I've seen it done quite well recently - unfortunately I can't find the animated hero right now, but the static part can be seen here.

Still, for some kinds of games, nothing beats handcrafted 2D sprites...

User avatar
Jetrel
Art Director
Posts: 7240
Joined: February 23rd, 2004, 3:36 am
Location: Midwest US

Re: Attracting and keeping artists on an OSS game project

Post by Jetrel » October 22nd, 2009, 11:16 pm

Absolutely: prerendered 3d graphics work very well.
Play Frogatto & Friends - a finished, open-source adventure game!

User avatar
Herduk
Posts: 97
Joined: August 18th, 2005, 9:19 am
Location: Bergamo - Italy

Re: Attracting and keeping artists on an OSS game project

Post by Herduk » December 2nd, 2009, 10:16 am

I got a question for Jetrel :)
First of all: i'm not an artist, i have hard time drawing stick-people :)
But i'm a good "manager".
Still i don't understand the difference between a "concept artist" and you "dictator-artist" to lead the art-part of a project.
You say (I simplify a lot, tell me if i misunderstand everything!) something like:
Concept artist = someone who tell the others how to uniform their style for the project
Dictator-artisti = a great artist that you put over the other to get an uniform style for the project.

I don't see too much differences between them.
So why the concept artist is bad and the dictator artist is good?
Don't bother a dwarf.. you can argue with his hammer!

User avatar
thespaceinvader
Retired Art Director
Posts: 8414
Joined: August 25th, 2007, 10:12 am
Location: Oxford, UK
Contact:

Re: Attracting and keeping artists on an OSS game project

Post by thespaceinvader » December 2nd, 2009, 10:26 am

A concept artist is someone who draws pictures which suggest how things should look (it could be that they're junior, not directly attached to the project, or someone senior without the requisite skills to do the remaining steps), providing usually a rough aesthetic, which other artists then interpret into game artwork. This is a completely unnecessary step when artists are at an incredible premium, as they are in the OSS community. You're better off just allowing the artist to create the game artwork directly.

Concept art is generally the first step in creating a look for a project or entity. They don't set the style immutably, they make suggestions for it.

However, you do need some direction. In order to have good artwork, there needs to be a coherent vision. When a game has only a single artist, this is easy - you simply let the artist have more or less free reign over the game art. When there are multiple collaborating artists (as with Wesnoth) you need one of them to be In Charge, or there will be competing visions and aesthetics, not to mention styles, and things get confused. You need someone to unify things.
http://thespaceinvader.co.uk | http://thespaceinvader.deviantart.com
Back to work. Current projects: Catching up on commits. Picking Meridia back up. Sprite animations, many and varied.

User avatar
pauxlo
Posts: 1024
Joined: September 19th, 2006, 8:54 pm

Re: Attracting and keeping artists on an OSS game project

Post by pauxlo » December 3rd, 2009, 4:11 pm

Herduk wrote: Still i don't understand the difference between a "concept artist" and you "dictator-artist" to lead the art-part of a project.
You say (I simplify a lot, tell me if i misunderstand everything!) something like:
Concept artist = someone who tell the others how to uniform their style for the project
Dictator-artist = a great artist that you put over the other to get an uniform style for the project.

I don't see too much differences between them.
So why the concept artist is bad and the dictator artist is good?
(The term is "director", not "dictator".)
The difference is that the director does also lots of work himself. Real work, not "concept work". And you usually get someone as your art director only when he has demonstrated some capability.

User avatar
Herduk
Posts: 97
Joined: August 18th, 2005, 9:19 am
Location: Bergamo - Italy

Re: Attracting and keeping artists on an OSS game project

Post by Herduk » December 4th, 2009, 8:51 am

pauxlo wrote: (The term is "director", not "dictator".)
I was referring to the first Jetrel's post
you pick the ones doing the most and best work (and who seem to have decent plans for getting the game done), and let them be dictators.
:wink:

Uhm.. i'm thinking about Invader and Jetrel say, but I still miss something.
Please, excuse me if i bother you with my questions, but i need to go deep in everything i find interesting, asking, sometimes, twice one or more points.

Ok, I understand the differences you point out from concept artist and the director/dictator.
I'm trying to understand now why an artist is more "attracted" by a projecte lead by a director/dictator than a concept artist.
At the end, a concept artist give you only few, very few, limits, for what i understand, while a director/dictator can be more strictly about that.
Don't bother a dwarf.. you can argue with his hammer!

User avatar
thespaceinvader
Retired Art Director
Posts: 8414
Joined: August 25th, 2007, 10:12 am
Location: Oxford, UK
Contact:

Re: Attracting and keeping artists on an OSS game project

Post by thespaceinvader » December 4th, 2009, 1:46 pm

That part is not about attracting artists, it's about keeping the ones you already have. You have to give them creative control (particularly your art director) or there is little incentive for them to keep going.
http://thespaceinvader.co.uk | http://thespaceinvader.deviantart.com
Back to work. Current projects: Catching up on commits. Picking Meridia back up. Sprite animations, many and varied.

User avatar
Jetrel
Art Director
Posts: 7240
Joined: February 23rd, 2004, 3:36 am
Location: Midwest US

Re: Attracting and keeping artists on an OSS game project

Post by Jetrel » February 24th, 2010, 9:42 am

Herduk wrote:
you pick the ones doing the most and best work (and who seem to have decent plans for getting the game done), and let them be dictators.
Ok, I understand the differences you point out from concept artist and the director/dictator.
I'm trying to understand now why an artist is more "attracted" by a projecte lead by a director/dictator than a concept artist.
At the end, a concept artist give you only few, very few, limits, for what i understand, while a director/dictator can be more strictly about that.
You're making the assumption that the project will have more than one artist; I'm making suggestions largely about single-man teams, because they're by far the most common. Even on most projects that have more perhaps 10 people contributing, there's usually one of them that does the lion's share of the work.

And that one deserves to makes all the decisions (about art). You give them that power, because you want to keep that person at all costs. It's extremely rare to have more than one such person; if you do, then they need to divvy authority between themselves, which is really just a matter of them getting along and being fair to each other. This is difficult and really isn't something I can make suggestions about, other than to avoid poisonous primadonnas and passive-aggressive types.


Really the crux of my suggestion is: if you land a great artist, don't try to control how they draw. Keep your opinions to yourself, and let them do whatever the hell they want, as long as they're creating resources that are actually useable in the game. Don't criticize their style, don't complain about their taste. Do crit technical stuff, but be gentle. Don't micromanage, because they're only coming to a free project to get away from that stuff.
Play Frogatto & Friends - a finished, open-source adventure game!

User avatar
Jetrel
Art Director
Posts: 7240
Joined: February 23rd, 2004, 3:36 am
Location: Midwest US

Re: Attracting and keeping artists on an OSS game project

Post by Jetrel » February 24th, 2010, 9:58 am

Also, a short note on pixel-art versus 3d art:

Pixel art matters because, given the right conditions, it's much easier to make a professional, good-looking bit of pixel art than it is to make an equivalent bit of 3d art. Easier in skills, easier in man-hours.

This holds true even for modest amounts of animation. When it stops holding true is when you have a huge number of assets, and you're trying to share stuff between them, or make very complex animations with lots of directions. Pixel-art absolutely shines for side-scrolling games, because you virtually have to make only one set of animations; there is no need to mirror stuff for multiple directions. It starts falling apart when you get into top-down or iso games; it's a good deal of work to provide the 3 sets necessary for N+S+E/W animations. It's a ton of work to provide the 5 sets for an 8-way (with mirroring) set of animations; generally beyond the capabilities of anything but a large professional team. Even an isometric game with 2 directions is twice as much work as a side-scroller.


The reason many genres, like RPGs and strategy games have switched to 3d is because 3d scales so much more nicely; you can reuse animations, you can get infinite numbers of facings for a unit (rather than the premade 4 or 8), you can tween animations to make them infinitely smooth, and last-but-not-least, you can even move the camera to a different angle, which is basically mandatory if you want to do anything cinematic.

The most clever end-of-the-era RPGs were able to dabble in cinematics by playing with the perspective of the background, but this only worked because their sprites were so tiny that it was hard to tell what angle they were drawn from (so much so that it had a delta of almost 45°). If you want decent resolution and cinematics, you basically get to choose one or the other if you're doing pixel art and want to not have to remake all your art assets all over again for the new angle.
Play Frogatto & Friends - a finished, open-source adventure game!

User avatar
Malvoisin
Posts: 22
Joined: February 27th, 2010, 1:43 pm
Contact:

Re: Attracting and keeping artists on an OSS game project

Post by Malvoisin » February 27th, 2010, 4:34 pm

To get back to the overall topic, I can only say that supplying a project with acceptable graphics can also be achieved by a different approach, i.e. skimming known artists' portfolios for their private work and email them about permission to use it for your game. I have done exactly that with our turn-based strategy game project called Birth of the Empires, and it worked better than I expected. I mailed about 200 artists and received over a fifty positive replies resulting in completely covering all graphical aspects of the game and some of them in nearly cutting-edge quality like the one provided by Aaron Sims for our second main trading race as you can see in the screenshot below.

I experienced that asking politely, being non-commercial, and of course having a somewhat fully playable pre-alpha version to download for the artist is enough to give a go for your project for quite a few artists. Of course you could say that using private portfolio work means people might recognize your graphics from somewhere else or have seen them already but most gamers don't look at portfolios in their spare time so that's negligible.
What could become more of a problem is integrating different graphical styles resulting from using artwork from many different artists into your game but I also experienced that all those 200 different artists I found all deliver very good and realistic graphics so depending on where and what you search there is a variety of cohesive and streamlined artwork available that way.

Besides it adds quite some to the fun when your credits read partially the same as the ones from Mass Effect when it comes to your lead artists 8) ;).

So I'd add another approach to getting good graphics and that is to dare asking around, even the more prominent artists. Our game got looking like this in the end (click to enlarge): Image

User avatar
thespaceinvader
Retired Art Director
Posts: 8414
Joined: August 25th, 2007, 10:12 am
Location: Oxford, UK
Contact:

Re: Attracting and keeping artists on an OSS game project

Post by thespaceinvader » February 27th, 2010, 4:44 pm

You can do it that way, sure. But you don't get the kind of cohesive look that we're developing for BfW. And it only really works for portraits and things. Getting sprites and animations done requires an intimate knowledge of the game engine.
http://thespaceinvader.co.uk | http://thespaceinvader.deviantart.com
Back to work. Current projects: Catching up on commits. Picking Meridia back up. Sprite animations, many and varied.

User avatar
Jetrel
Art Director
Posts: 7240
Joined: February 23rd, 2004, 3:36 am
Location: Midwest US

Re: Attracting and keeping artists on an OSS game project

Post by Jetrel » March 4th, 2010, 8:45 am

thespaceinvader wrote:You can do it that way, sure. But you don't get the kind of cohesive look that we're developing for BfW. And it only really works for portraits and things. Getting sprites and animations done requires an intimate knowledge of the game engine.
Yeah. That doesn't work for 3d models and sprites.
Play Frogatto & Friends - a finished, open-source adventure game!

User avatar
Malvoisin
Posts: 22
Joined: February 27th, 2010, 1:43 pm
Contact:

Re: Attracting and keeping artists on an OSS game project

Post by Malvoisin » March 4th, 2010, 9:35 am

maybe not all that strictly I think. Just have a look at Space Empires (IV) for example. Doing so you'll notice most of their ship graphics come from outside artists and modders and these are complete sets of 20+ ships each featuring a pretty cohesive look for each single race they are developed there IMHO. Now Atrocities, a major modder there, for example has created all his pictures from shots of low-poly (kind of game-ready) DOGA 3D models he created. He lent us these shots in large resolution and also the 3d models natively ported to .X once our combat engine is in place. .X is pretty much standard for 3d models, at least it's not so exotic that you could not write an engine for.

I think it's all the question of artwork being developed for the right purpose and scope that the artist intended in the first place, for example personal portfolio showcase stuff tends mostly to be low in numbers (that's why I had to mail a lot of artists for the race pictures) while an artist creating a whole personal world in space needs to do a lot more models that are much more cohesive, for example Aaron Sims who did just that with Tethered Islands, a personal project of his that he does for fun and to write books and sell learning DVDs about how he created it, yet of course retaining all copyrights himself as long as he does not sell it to a game or video company. This makes it possible for a project like ours to just ask him about usage rights.

In case of Atrocities I had just to mail once and got all 100+ completely original design ship models and pictures I needed (except the minor race ones, but they can for now be neglected) ;). The only thing that might be an issue here is if the genre is too close and people playing our game too often recognize the same graphics that they know from Space Empires. But there's always a downside in everything, isn't it ;).

It's a bit unusual approach I admit, feeding oneself from artwork done for private world building purposes or for games somewhat similar to your own game, at least also setting in space which is here the main common denominator.

For sprites (explosion animations on the map etc. which 3d models would be too much for or used to create sprite sets) I haven't looked around yet. Though we have a self-coded working sprite engine we haven't felt the need to include sprites yet.

User avatar
Spell
Posts: 48
Joined: January 23rd, 2009, 9:04 pm
Contact:

Re: Attracting and keeping artists on an OSS game project

Post by Spell » April 11th, 2010, 11:54 am

I dont know if someone mentioned it before:

You need usable graphics from the beginning. They don't need to be really great but atleast shouldn't look like [censored]. S****y Horrible graphics scare away all the players + the artists. All better open source games I've seen had quite usable graphics from beginning. If the game has s****y horrible graphics, no matter how good the gameplay is, the people don't even look at it. As thus atleast one of main developers must have artistic talent or the game won't have any success. If you are not good at graphicsmaking you should try some abstract style of graphics, for example stickmen.

Edit: Damn censorship.

Post Reply