Observer:Profiles

Transcript: An Interview With Sapient

Hi, Sapient, and welcome to The Observer. We have five questions we would like to ask you, today.

1) How did you find wesnoth, and why have you stayed with the project for two years? What pull keeps you here, Sapient?

I found wesnoth from a long and winding path. I used to hang out a lot on Radio Rivendell forums, then one day a guy posted a link in there to some video game music. The website led over to some free video game reviews, then that led over to Wesnoth. I was a Wesnoth player a long time before I touched the source code. I played around a lot with WML and had a lot of fun with it, As you can see from my old map pack, which still gets a lot of downloads I see. Then my relationship with Wesnoth changed a lot when I started getting into the C++. At first I just wanted to improve the MP Timer controls, then Soliton talked me into doing some UI improvements. He's a real good proponent for the project. I never read a book on C++ or officially studied it in a class, but I knew a lot about C and Java. It was fun learning the new language and I enjoyed it a lot; Xan was a good tutor when I had a question. I've never actually completed an official campaign to this day, unless you count the tutorial, So I guess you could say I'm a bit odd in that I like coding more than playing. I used to actually be addicted to video games such as MMORPGS so I try to keep a tight tab on how much time I play

2) You dedicated to improving the WML. Are you coding in any new features at this moment? Is it difficult to manipulate WML at the source level?

Yes, as you have noticed improving WML is one of the areas that I find the most fun. I have more plans for improving the language as well. Currently, I'd like to give the WML author a way to store all locations that are reachable by certain unit(s), and a way to use the pathfinding algorithm to store the path a unit would take if it can go from A to B. As for crazy ideas, maybe WML should have support for functions/calling. I recently refactored the Standard Location Filter code, but that was an internal change. The WML authors won't notice it, just C++ coders. Before that I made the C++ for WML variable lookups / modifications much cleaner. Again, an internal change; but one that will help future coders. These were difficult changes, yes, but they will make future work easier. As for creating WML events/additions, some of them are very very simple. But it requires a lot of foresight and planning so that you are adding value to the language and you don't have to go back later to change the way something works. I remember from my days as a WML Author that losing backwards compatibility is a real nuisance and a deterrent to development.

3) Have you ever been tempted to misuse your power? (even if you have not, in the end)

Honestly, no; I'm a pretty humble soul. I never wanted to give myself uber powers through manipulations. Although I do try to think up such scenarios sometimes. I think like a criminal mastermind sometimes, but this is more to put myself in someone else's frame of mind. Then you can anticipate their moves.

4) What sort of computer do you use, and what OS? Why do you use it?

That's an interesting question. Maxy used to be the dev with the lowest end computer, but now I'm pretty sure I have that distinction. I'm running Windows XP Professional, on a Pentium 4 with 256 MB Ram. I have a 16MB video card. I intentionally had them downgrade it from a 32MB video card when I ordered the laptop just so I could get more work done in college and graduate. (Let that be a lesson to you college students, heh!) As for why I use this OS, well I'm a creature of habit I guess. I got everything set up the way I want it over a process of many years and I don't have enough drive space or spare computers or backup devices to really experiment too much with it. One of these days, I'll do that, at least, that's what I keep telling myself anyways.

5) Sapient, does your dodging ninja avatar on the forums have any deeper meaning? What does it represent, symbolically?

Yes, my avatar does have a deeper meaning. I'm not sure I want to give it away, though! I used to play a lot as Northerners and I would have some really lucky assassin strategies sometimes - I developed the saying "You can't kill a true ninja!" I have an appreciation for bounty hunter fiction stories, too. I was once contemplating creating a campaign based on the story of a bounty hunter during the beginning of Ashveire's reign. He would basically be tasked with rounding up dissidents who were still loyal to the previous ruler. And the game play would be focused on using the fewest number of units possible, rather than the usual bonuses.

Transcript: An Interview With Jetryl

1) As the art director for Wesnoth, where do you plan to take us, graphically? Do you have any 'big project' under way at the moment?

First, I intend to see the team-coloration project to fruition; to give the drakes, naga, and mermen the same recolorable patches (and general cleanups to their sprites) that all the other races have.

After that, there are a few possibilities - completing these would put us above the graphical level of quality seen in commercial games like final fantasy tactics, and because our graphics could (and have already) made other, radically different games possible, it would be a win for OSS as a whole. These include: - Giving our units a set of northeast facings similar to our current southeast facings. This would solve most of the graphical cheapness that happens when one of our units plays an animation that depicts him swinging his weapon in the opposite direction of his opponent.

- Giving our units walking animations. Far more difficult than the job I just mentioned, but this could well launch our content into being suitable for an RTS - something which open source has already finished code for in the form of the Stratagus Engine. It would also make Wesnoth itself look incredibly good - there are many similar games like FFT, Disgaea, and Tactics Ogre that demonstrate that walking animations work in a game of our type, and greatly enhance the visual experience. When the time comes, I'll probably be leveraging a lot of community help on this.

- Providing our game with a single set of high quality, consistent portraits.

- Cleaning up a few loose ends in the terrain department - coming up with a better solution to dirt tiles/transitions and snow transitions. Coming up with better bridges, replacing the desert hill/mountain tiles, and also finishing a start that freim made on snow mountain tiles.

2) When you first joined, nearly four years ago, the art was quite different than it is now. How do you feel about the changes since then? Do you ever want to go back, and play the pre 1.0 releases?

I really don't have any nostalgia for the old graphics, mostly because (at least ideally) the new graphics have preserved everything that was good about the old ones, and discarded everything that was bad.

If nothing else; looking at the old versions really offers a sense of accomplishment. I feel that we've made as much of an improvement between the days of 0.6 and 1.3.x as was made between many sequels of commercial games. We still have plenty of things to do, but what has been done so far is very encouraging.

3) What is your favorite program to draw with?

Photoshop. It's a good "jack of all trades" program, and I've found it strangely better at pixel art that a lot of the programs out there that are dedicated solely to making pixel art easier. I think a lot of this has to do with depth in other features that photoshop developed to make general image editing easier, but which are equally applicable to sprite art and which are missing from a lot of the simpler "made for sprite art" programs.

These include things like photoshop's extensive history support, not only being able to "undo" as far as 50 steps back, but also being able to "save the state of the document", so that you can explore a certain direction with a work, but are still able to roll back to the old one no matter how much further you progress into the work. The 800lb gorilla in the room is probably layers - I can place elements which would obscure things behind them, but which I'm unsure of the right position of, in a separate layer. For example, I can put the shield of a sprite in one layer, and his mostly-complete body underneath. As the balance of the sprite's pose changes during the drawing, I can shift his shield to accommodate with almost no drawing whatsoever. Photoshop's color manipulation tools are also useful for quick previews of radically different color schemes.

The GIMP is another program that a number of our artists use. I'm not terribly fond of it's interface, since it makes certain basic operations that I do in photoshop take more time (such as selecting a range of pixels and translating their position without cutting and pasting them). I believe a few of these things came from an attempt to better position the gimp as a "photo editor" than Photoshop, which has some bias in its interface conventions that it inherited from older programs like Silicon Beach's SuperPaint - these programs had an interface for basic editing (drawing, selecting, moving) that was more like what the program MS Paint now has. However, I suspect that this might just be my acclimation to photoshop's style of interface, and I know that the gimp can be very powerful when a person becomes experienced at using it. It also has the benefit of being "free software" in both senses of the term.

4) What do you think of the new animations that the 1.3.x line introduced?

The centerpiece of the new animations has been the idle animations, and I think they really help with a major flaw that wesnoth used to, and still does have. In many older games, even games within our general category, there is a considerable amount of visual action on the screen even when no gameplay interaction is taking place. I strongly suspect that this visual motion, this "liveliness" to the whole screen has a considerable and natural psychological appeal to players; in much the same way that vibrant, lively colors do. If you look even at the classic game civilization, it had animation of its water terrain through palette cycling, which often resulted in a while third or more of the screen having some sense of motion to it. Simcity (especially 2000 and thereafter), went to the length of animating many of the building tiles. Our idle animations don't completely solve this problem, but they do a lot to help with it.

Boucman, as well, has done a number of changes and major cleanups with our animation WML which will really ease a number of future additions; mostly what I mentioned in the answer to question #1. He, and zookeeper who helped him with the content end, really deserve a standing ovation; they're the people who really made these improvements possible.

5) Is it difficult to deal with the different art styles brought to wesnoth? How do you reconcile the different artist's work to produce a 'Wesnoth' style?

It is difficult, and it's resulted in a lot of wasted work. By far, the leading cause of "lost work" has been poor work that wasn't up to snuff, rather than bits of good work done in separate styles. Our sprite art, our terrain art, and our attack icons have all stabilized in style, and no one that was good enough to do them the right way has been so wasteful as to make images blatantly in the wrong style. Our portraits have not quite stabilized in style; we have more than one body of work in disparate "styles", and this is a problem we'll need to resolve to get a full set (consider the difference between the drake and saurian portraits, and the human portraits).

This lack of quality would be an issue with a commercial company, but they deal with it in a different way - they simply end or never begin an artist's employment. I have to make the same judgement that a commercial company would make, but on a piece by piece basis - often I can work with someone better to bring them up to par (like leonhard, who did some great work on the attack icons), but even with my own work I have to carefully judge if a piece is of the right quality to fit in.

One of the problems we've run into is that it's far more attractive to most players to try to expand our content horizontally, than it is to expand it vertically. It's far more appealing to try and make a new monster, than to make the same number of images for an existing unit to improve its animations. A lot of this, I think, has to do with "authorship" - when someone makes a whole new monster, they can call it their own. The upside of this is that this phenomenon can be turned to our ends, but only if the person is willing to do enough work that whole units' worth of images can be done by them. People like Eternal and Neoriceisgood have done some excellent work on our team-color redesign, and have probably been spurred by the fact that by doing all the work necessary for the new imagery of a unit, they could really call it their own. We've had success by giving people like that "artistic license" over a major area if they're willing to do all the relevant work; I've certainly been a bit motivated by this myself. It's darkly ironic, because if the people who do UMC were willing to work on core content, their work would be seen every time someone played wesnoth - far more often than it would on UMC, and vastly more so than on content that never got finished; that was just left to die in a forum thread for some pipe-dream campaign/era. I think that a lot of this phenomenon is similar to what ESR described here: http://www.firstmonday.org/issues/issue3_10/raymond/index.html

How I reconcile the art is simple - if it's even directed at something useful to begin with, I make a judgement of whether it would be economical for the game if I spent my time editing it into shape. Sometimes something is far enough off-base that my time is simply better spent recreating it from scratch. At those times I either do exactly that, or just ignore the piece and continue working on what I was doing at the time. A lot of the time, I try to offer advice to people on how to improve their work - it's actually spurred a number of people to continue helping for quite a while, and has improved some to the point where they're independently able to do mainline work. The mechanics of how I actually fix an individual piece are probably beyond the scope of an interview like this - I'd probably have to explain a lot of "how to draw". Return to Page 1