Interview with Jaakko Tapani Peltonen of NetHack: Falcon's Eye

Matt Barton's picture

Falcon's Eye: NetHack + Graphics = Awesome!Falcon's Eye: NetHack + Graphics = Awesome!I recently had the pleasure of talking to Jaakko, one of the developers of the ongoing free software project NetHack: Falcon's Eye, an awesome, graphical remake of the classic Rogue-style game NetHack. The game is available for Windows, Linux, and DOS platforms. The best part is--it's FREE. Below are some questions that Jaakko was kind enough to answer about Falcon's Eye. It's in-depth and makes for some very good reading, especially for those interested in classic CRPGs. Enjoy!

I recently talked with Hajo Malthaner about his project Iso-Angband, which unfortunately does not seem to have fared very well. Malthaner argues that fans of Roguelikes simply don't like the idea of graphics and were resistant. Do you have the same problem with your game's reception? Is the community just not happy about the idea?

My experience with NetHack - Falcon's Eye has been different; I feel there has been a lot of interest. In more concrete terms, there have been over 178000 downloads on of the Windows version alone (over 238000 of all packages in total), and the game has even been included in some Linux distributions. I've personally received over a thousand emails about the project.

I don't think there is a single "community opinion" about graphics. In particular, long-time roguelike fans and casual players may have different tastes.

Certainly there are many who prefer text-mode interfaces (or simple graphical tiles). They have some concrete advantages: for example, as often mentioned, they show a whole dungeon level at once, which large graphics cannot do. It is also a matter of taste: long-time roguelike fans might be used to text-mode interfaces.

On the other hand, for people who discover roguelikes nowadays, non-graphical roguelikes may seem disappointing compared to commercial games, despite the complexity of the underlying gameplay. Graphical interfaces can be appealing to such people and thus can introduce new players to the genre: many people who have played NetHack - Falcon's Eye have apparently discovered or rediscovered NetHack through the new graphical look.

Note, though, that in NetHack - Falcon's Eye the interface does more than simply use a graphical tileset; there's music, sounds, and a point-and-click interface instead of the typical keyboard interface. Maybe that helps make the result interesting to people who don't see enough benefits in graphical tiles alone.

In an age of highly sophisticated CRPGs, it's surprising that the roguelikes are still so widely played and enjoyed today. What is about these games that continues to appeal to modern gamers?

From one viewpoint, I think roguelikes represent a kind of distilled CRPG gameplay. Modern CRPGs use a lot of scripted sequences and plot exposition; the benefits for storytelling are obvious, but the drawback is that part of the playing time is spent on setting up the situation, and the player must follow along. By comparison, in roguelikes plot, dialogue, puzzles and character interaction are minimal but the actual gameplay of dungeon-delving is in many ways just as sophisticated as in other CRPGs. I think the only significant gameplay feature missing in roguelikes is having a party of multiple characters. This kind of "distilled to the essentials" gameplay can be very appealing.

It's very interesting that non-graphical roguelikes continue to do so well even though non-graphical games in other genres are almost nonexistent. About the only other exception I can think of is the
"interactive fiction" (text adventure) genre, but maybe I have missed some others. Arguably, CRPG gameplay might be less dependent on graphics than many other genres; after all, gameplay in pen-and-paper RPGs is not graphics-based at all.

One thing that I think may be common to many CRPG players is the wish to not only play the games but also to somehow make their own roleplaying game - after all, pen-and-paper RPGs are all about
playing your own stories. Roguelikes are an answer to that wish, in the sense that one person (or a few) can feasibly create a complete a roguelike from scratch, without requiring years of development time or large amounts of graphics and audio creation.

So this do-it-yourself aspect probably generates some interest in roguelikes; both for the actual developers and for players who can feel like they are playing custom-made games. On the other hand,
recent mod-friendly commercial games provide another outlet for do-it-yourself CRPG players; it will be interesting to see whether this affects roguelike playing or development.

How long have you been playing roguelikes? Can you talk about your earliest exposure to this genre, and the games you played? Why "Nethack" instead of one of the other roguelikes?

I am not a harcore roguelike player, but more of a casual player. I have played a large amount of CRPGs (ranging from old Ultimas and Gold Box AD&D games to Baldurs Gates and Elder Scrolls games), but I was never that much interested in roguelikes.

Isometric View: Try it for yourself! It's free!Isometric View: Try it for yourself! It's free!In fact, the work that led to NetHack - Falcon's Eye was, at the beginning, not related to roguelikes at all. I was working on an isometric graphics system for fun, a precursor of the one used in NetHack - Falcon's Eye. Making a whole game to go with the graphics system was too large a task, though. Early on I had a tentative idea of using the graphics system to make a graphical upgrade for an old CRPG masterpiece, Ultima IV, but I did not pursue that.

I soon looked for another game whose graphics I could upgrade, and of course, that is much easier to do for a game with source code available. I don't remember whether I looked for roguelikes especially, but eventually I thought of NetHack.

Why I thought of NetHack is a mixture of several reasons. It is one of the more well-known roguelikes, and I have also seen it mentioned occassionally in the Finnish gaming magazine "Pelit" (they have even reviewed it twice), so you could say NetHack had good publicity. I had also played NetHack before - I remember having it on a 3.5 inch diskette - but I don't recall my first encounter with it anymore.

NetHack is also technically well structured so that creating a graphical interface is reasonably easy - I just had to write new versions of a few key parts, instead of making changes throughout the source code. That made it possible to finish the project without becoming an expert on all aspects of the code. I think this would not have been so easy in some other roguelikes.

At least later on, with NetHack - Falcon's Eye already released, I had considered making graphical upgrades for other roguelikes too. In particular, Angband interested me, but I was concerned about its non-opensource license and about possible copyright issues with the Tolkien influences. I did create an Angband title screen graphic for fun, but that's as far as it went.

Nowadays I have of course played roguelikes much more, both while testing my changes and just for fun. You could say that, through the NetHack - Falcon's Eye project, I made it easier for myself too to get into roguelikes.

Interestingly, the original Rogue was never a commercial success even back in the early 1980s. Why do you think roguelikes haven't appealed more to the average gamer? Is the love of the game just somehow tied up with the UNIX and GNU/Linux communities?

There are historical reasons for the UNIX or GNU/Linux connection; free (open source) software development coincides well with developing free games. However, I don't think the attraction of roguelikes is in any way limited to UNIX/Linux users. In fact, there have been successful commercial games that are at least related to roguelikes; the most well-known are likely the Diablo games, and there have been many similar ones since, e.g. the Sacred games, Titan Quest, and others. 3D dungeon crawling games like Dungeon Master and Eye of the Beholder were also rather similar to roguelikes, which was most apparent in "Dungeon Hack".

The lack of graphics and sound certainly hurts initial perception of the traditional roguelikes. Screenshots are probably among the first things one looks at when one learns about a new game. If the visuals and audio seem poor, few gamers will stay interested long enough to download the game, or play long enough to discover the fun in the underlying gameplay, especially since they may have better-looking CRPGs to spend time on.

Traditional roguelikes can also have a high learning curve - there are no in-game tutorials, and many things can only be learned through accident, experiments, or by reading "spoilers" on the internet. Moreover, the games are very challenging, not least because there often is no real savegame feature, so players cannot retry failures except by backing up some files outside the game. Average gamers might not enjoy "Yet Another Stupid Death" requiring them to restart.

The turn-based gameplay might be a problem for some gamers; turn-based games have been on the decline for many years now, except for strategy games. Even Dungeon Master and Eye of the Beholder already used a kind of "partly real-time" combat, and Diablo was, at least to the player, fully real-time. The Fallouts (and maybe Arcanum) were probably the last well-known turn-based CRPGs. I think turn-based gameplay hurts single-character games more than party-based ones where the tactical
aspect is a good justification for turn-based combat. Still, I don't think this is a crucial disadvantage for attracting average gamers.

So what should be done if one wanted to make roguelikes more appealing to average gamers, or even commercially appealing?

To lower the learning curve, I think savegames should be allowed in standard roguelike gameplay. This does not mean discarding the traditional "see how far you can get" play type: playing without reloading could be a voluntary challenge that the game keeps track of, much like NetHack keeps track of various other conducts.

A commercial roguelike would almost certainly add graphics, sound (as a well-supported option at least), as well as tutorials and savegames. The downsides of large graphics can, I think, be dealt with; for example the issue of seeing the whole dungeon level could be addressed by map displays, like I've tried to do in NetHack -Falcon's Eye.

For free roguelikes this would of course mean a lot of work creating graphics and sounds. Perhaps there could be some common repository of free roguelike graphics and sounds to draw from; while there would still be problems related to mismatching graphics styles, tile scales and orientations, etc., it might still reduce the workload.

I don't think there's much more that needs to be done; of course, making the games known by publicizing information about them and making good websites is always useful, but how to do that well
is a more generic issue affecting all small game projects, commercial or free.

What are your long-term plans for your game?

I've been involved with NetHack - Falcon's Eye for a long time. It has certainly been a learning experience: it was my first big DirectX application, and later became my first Linux application, my first MacOS X application, and even my first BeOS application. When I released the first version in 2000,
I was studying for my master's degree. Now I am a postdoctoral researcher, and I find I have less free time than I used to, due both to work and to other projects I work on. As a result, although I have kept on answering emails and message forum posts about the game, actual game updates from me have been mostly on hold for some time.

I am somewhat ashamed about this - at times I've said I intend to release a new version over some span of time, yet that has always failed. On the other hand, it's very gratifying to notice how much people remain interested in the game; in particular, I should mention the hard work of the people in the "Vulture's" project, who started a "fork" of the game to maintain ongoing development.

Despite the constantly slipping update deadlines, I still very much intend to release new versions. I want to add more graphics: for example, maybe a different tile orientation than the current 45-degree view, and portraits for the description windows about different monsters and items. The variety of
sounds and music should also be much expanded. I also want to make it easier for other people to participate by adding new graphics.

Over a longer period, though, I think I will need to look for additional people to take over the main duties of maintaining development. That doesn't mean I would stop development completely myself, but my contributions would become less frequent (as they already are).


Bill Loguidice
Bill Loguidice's picture
Joined: 12/31/1969
Super NetHack

The graphics and whole overhaul concept are very well done and the fact that he did all this for free is even more impressive. Great stuff. I hope to someday get to try it.

Bill Loguidice, Managing Director
Armchair Arcade, Inc.
(A PC Magazine Top 100 Website)


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.