Interview with Rogueknight333: Author of the Swordflight Series for Neverwinter Nights

Rogueknight333

In case my 25-part coverage has not made it clear, by reason of its reactivity, resource management and tactical combat, along with its replayability, attention to detail and overall polish, I hold Rogueknight333's Swordflight in especially high regard. In fact, to my mind, Swordflight is not only the best single-player campaign to grace the Aurora platform but also one of the best cRPGs ever, amateur or pro.

With the imminent release of the highly-anticipated fourth installment of Swordflight (currently in beta form and being playtested), Rogueknight333 has kindly agreed to my written interview request. So without further ado, let's get to it!

If Swordflight's design is any indication, you seem to have played your fair share of oldschool cRPGs, strategy games and text-based or point-&-click adventure games. So, I'd like to know, what games of the past do you remember most fondly and which ones (if any) inspired you in your development of Swordflight?

I have certainly played a fair share of various sort of games (though there are also some significant games I have missed as well – time is limited), but the Infinity Engine games (Baldur's Gate 1+2, Icewind Dale 1+2, Planescape: Torment) are both particular favorites of mine and significant influences on Swordflight. The Baldur's Gate games in particular are surely the single biggest influence, as is probably rather obvious to anyone who has played both series. Another much less obvious game that I would rate as a significant influence on my own game designing is Deathlord. I played this long ago (the game was released in 1987 IIRC) on a Commodore 64 and have remembered it ever since for how brutally and often unfairly difficult it was. It was not necessarily that good a game on balance (even by the standards of the time), but it managed to make a lasting impression by being so hard. It was just so satisfying when you actually managed to clear a dungeon or otherwise accomplish something significant. I have tried to give Swordflight at least a touch of the sadistic cruelty that (apparently) went into designing Deathlord.

Swordflight feels eerily similar to PnP/tabletop D&D. Have you ever played/DMed in the traditional way? I don't mean online via forums or email but rather face-to-face with friends, drinking beer and munching on pretzels.

Perhaps it is my muse rather than myself who should be credited for that eerie similarity, because my PnP experience is actually quite limited. I have rarely had more than one or two friends available for playing tabletop games (not really enough for a proper D&D session), and we have more often played historically based wargames rather than RPGs. Possibly my tendency to emphasize tactical combat can be partly traced to that background.

When turning to computer games, however, I found the historical wargames in that medium tended to be far too easy due to the weak AIs, so I ultimately ended up preferring other types of games (such as RPGs) in that context, where the more artificial situations made the AIs better able to provide a reasonable challenge. This is in part simply because a lack of realism makes it easier to give the computer opponent unfair advantages, though in my experience, AIs tend to be effective in inverse proportion to how realistic the scenario they are confronting is (e.g., very good at playing pure abstract games like Chess, absolutely terrible at realistic historical simulations).

Why did you choose to create a campaign set in FRCS rather than in something like Dark Sun or your own custom world?

It was a setting I happened to be familiar with and that I could expect many players to be familiar with as well. This limited the amount of background research (or writing for a custom setting) and lore dumps that would be necessary. It was also a setting that the NWN engine and base content (originally designed around a FR campaign) naturally lent itself to, with limited need for infusions of custom content. All of this saved time and simplified design, which are important things to do if a 100+ hour epic RPG is going to be made by one guy in his spare time.

The Aurora platform hosts hundreds of single-player adventure modules and campaigns, many of which predate Swordflight. Which of these (if any) could be considered "precursors" to Swordflight? Also, what did they do right and what did they do wrong? 

SteveB's Blackguard trilogy and Yaballa's Sapphire Star are among the most challenging combat-oriented modules. They are similar to Swordflight in this respect, and were helpful in convincing me that it was possible to make interesting tactical combat in NWN.

Rick Burton's Paladin trilogy (Twilight and Midnight – the intended third part was never made) includes high quality modules whose basic plot, with the PC acting as a mentor and guard of sorts to a special henchman companion, somewhat anticipates that of Swordflight in certain ways.

What became the Witch's Wake premium module (a version of which had actually been released by Bioware as a free download in the early days of NWN, some time before the premium module program got started) did an excellent job of demonstrating the Aurora toolset's ability to make a campaign with highly reactive role-playing, though unfortunately its brevity and unfinished state makes this module less than satisfactory as anything other than a demo.

Baldecaran's early modules, Cave of Songs and Honor Among Thieves also greatly impressed me with the capabilities of the Aurora toolset.

There are of course many excellent NWN modules that either had not come out or that I had not gotten around to playing until I was well into the making of Swordflight, and thus cannot be considered precursors to or influences on it, without that reflecting any discredit on them.

The best NWN modules tend to be quite good at whatever they are primarily focused on, but often are too specialized to make good all-around RPGs. In general the NWN modules that are best at story-telling tend to be rather weak when it comes to combat design, or just not have that much combat, and the NWN modules that have decent combat tend to be deficient when it comes to the writing, role-playing reactivity, etc. There are also far too few of these latter.

Prior to Aurora, did you have any experience in modding and using game editors?

I would on occasion partially rewrite the rules of some of the board wargames I played to make them more historically accurate. With computer games, I had often played around with various game editors a bit, but to little result, with one notable exception. This was a longish campaign I made for Blizzard's Starcraft (a Real-Time Strategy game which I was playing a lot in the late 1990s), using that game's Editor. It was complete and playable, though in need of some fine tuning and balance adjustments, but never publicly released anywhere, since I mostly just made it for my own amusement and as a sort of training ground for the competitive multi-player, which was the main focus of Starcraft. I am not sure how good it was, but I think the practice of writing a complete story with a beginning, middle and end, in a gaming medium and combined with challenging combat scenarios, was a helpful experience that has served me well in making Swordflight, despite these being extremely different types of game.

Like Swordflight that campaign could also be extremely difficult. Perhaps I should play it again some time and see if I am actually capable of completing it now that I am getting a bit too old and slow for full-fledged real time games.

You share with certain other veteran builders are marked distaste for the Electron engine which powers Neverwinter Nights 2. What are the main reasons for your decision not to build Swordflight for Electron? 

There are numerous problems with the Electron toolset, many of which would be small and nit-picky by themselves, but a lot of small problems can add up. There is however one huge and dominating problem, namely Electron's use of sculpted terrain rather than tilesets. This makes building an area in Electron take far too long. In Aurora, I can make the basic terrain and outlines of an area (excluding the design of encounters, writing dialogue, special scripts, etc.), in 15 minutes or less. The same task in Electron would take several hours at a minimum, possibly longer. For an amateur modder working by himself in his limited spare time, this is a serious problem. Making efficient use of time is absolutely essential to a project like Swordflight. Saving time is, indeed, the chief point of using a toolset or game editor in the first place, for if time were unlimited it would make more sense to just make a full-fledged indie game from scratch. This cost does not even come with much in the way of benefits, since if one makes extensive use of custom tilesets one can make areas in NWN that look almost as good if not better than NWN2 areas.

Aside from the issues with the toolset, I find NWN2 to be just a worse game than NWN (in terms of the basic game systems not the official campaign content of course – in the latter NWN2 and especially MotB is much superior to NWN). It is full of bugs, over-complicated kludgy systems, and terribly balanced classes. I just did not find it that much fun to play, which naturally did not help motivate me to build modules for it.

Overall, NWN2, both toolset and game, is a cautionary tale about making something based on a laundry list of features people think they want without regard for how or whether those features fit together in any sort of cohesive whole.

What did you set out to achieve in Swordflight? Was your ambition to build a Baldur's Gate-eque campaign from the get-go or did Swordflight just grow into such through its development?

Primarily I was just trying to make the same sort of RPG that I enjoyed playing. I was also trying to make a series that would be more well-rounded than the typical NWN module, with good combat, role-playing reactivity and story-telling. I am not sure how well I succeeded at that, as most commentary on it, both positive and negative, has tended to emphasize the combat.

I always intended Swordflight to be a long Baldur's Gate-esque epic. From the beginning I had a general outline of the overall story in mind, though most of the side-quests and such were things I made up as I went along. The planned scale of the series was probably foolish, as it would have been much easier to start with a smaller module. In the event, the series grew beyond even my already excessive ambitions in the making. So far, every chapter except the first has ended up being much bigger than I originally planned.

On the new Vault, the reception to Swordflight has been decidedly positive; this, despite its difficulty and demand for player attentiveness and time investment (dedication) in comparison to other campaigns of its kind, but how was Swordflight received on the old IGN Vault back in 2008?

Swordflight's initial reception could be described as mixed at best. Most of the people who first played it seemed rather shocked by the level of difficulty, which was certainly noticeably greater than the average for NWN modules. I received little in the way of 10 votes and unqualified praise. Now I certainly do not think it was perfect or deserving of unqualified praise, but, unfortunately, given the voting inflation that prevailed on the old vault, a lack of such all but guaranteed that the module would sink into obscurity soon after release, which duly happened. I almost gave up on continuing the series at that point, as it seemed there was little interest in the kind of modules I wanted to make. However, the first chapter was eventually given a very positive review by QSW aka QueenSilverwing of the old vault's Reviewer's Guild, which helped call attention to it and generate a download count that was not absolutely terrible. It was clear by that point that I was building for something of a niche audience, but now that more people were going in with some warning about what to expect the comments and votes on average tended to be more positive. When Ch. 2 was released most of the feedback was positive, there just was not very much of it, due to how late in NWN's history it came out. However, that lateness probably worked out to my advantage on the new vault, where the Swordflight modules, relatively speaking, have gotten a lot of attention. I suspect this is in large part simply because I am one of the few NWN authors still active.

You could be referred to as a champion of resource management, rest restrictions and overall dungeon strategy. These are oldschool principles of RPG design that current gen devs seem reluctant to employ. Why is that?

There are numerous advantages to restricting resources and requiring careful management of them. Aside from generally increasing the challenge, it adds an element of strategy to the game, as one must plan for multiple encounters rather than treating each as entirely self-contained, it can make even trashmob encounters meaningful (you might win easily, but you still might have to expend a limited resource to do it), aids class balance (a class weaker in a particular encounter can still beat it by expending more resources), and makes it easier to provide meaningful loot (providing a powerful but limited use item can make a real difference without altering the game balance for the rest of the game).

It seems to me that many if not most of the problems with current gen games stem from the fact that, at some point, game devs became afraid to make games that are actually hard. As to why they developed this aversion, I do not think I have any special insight, and can only speculate.

I suspect that part of the problem is that gaming has become a victim of its own success. When I first started playing cRPGs, it was assumed that the people playing them were serious gamers willing to devote time and effort to mastering difficult games. Nowadays, by contrast, gaming is a much more mainstream recreational activity, with far more casual players who just want to relax with a game every now and then but not make it a big part of their lives.

I would also hypothesize that the emergence of easy and widespread piracy in the internet era has hurt the quality of games. A huge AAA game aimed at the widest possible audience can still make a profit in such an environment, but products aimed at a more niche audience will have a harder time. This has also encouraged game companies to give more emphasis to MMOs and console games (with the former depending on subscriptions and the latter on physical media for a good part of their profits, piracy is less of a factor), with unfortunate consequences for standard cRPGs.

Whatever the reasons (the above are basically just guesses), I fear my own experience with Swordflight's poor initial reception makes it hard to conclude that game devs are mistaken, from their own point of view, in seeing easy, dumbed-down games as the way to go. Had Swordflight been a commercial product, it would have been a spectacular, money-losing failure.

Swordflight is a highly reactive campaign. Why is reactivity important in cRPGs?

Role-playing calls for NPCs and the game world generally to actually react to the role the PC has taken. To the extent that there is no real difference between playing a game as one type of character rather than another, if everyone reacts to a dwarf in the same way they react to an elf, and to a barbarian exactly as they do to wizard, and so on, the choice of one over the other has become meaningless. Thus, reactivity on the part of the game world is the appropriate complement to role-playing on the part of the player.

Another note to make here is that combat difficulty contributes to reactivity, since combat that is actually hard requires one to make full use of the specific capabilities of particular characters. By contrast, if it is very easy one can play, for example, a wizard who never casts spells and just pretends to be a fighter, yet still wins.

Other than gradually influencing Zarala's alignment over the course of the campaign (possibly for a build up towards some kind of grand event that takes her alignment into account), will there be any other examples of far-reaching, perhaps Chapter-spanning reactivity in Swordflight? (For example, choices made in Chapter Two have consequences in Chapter Six.)

There are a few early chapter events whose outcomes were noted in the database so as to possibly have consequences later, but not too many. I am not sure how well players are actually preserving the databases and characters during the long waits between chapters, so I am hesitant to make anything too important depend on them.

As rough estimates of Chapter length, you have Chapter One (2008) at 12 hours, Chapter Two (2012) at 50 hours and Chapter Three (2016) at 30 hours. What is your rough estimate for the upcoming Chapter Four?

I expect it to be similar in length to Ch. 3, 30 hours or so.

Also, my commentary has (crudely) simplified the nature of Chapters 1-3 as follows:

◦ Ch. 1: A difficult and very well polished multi-level dungeon (Buried Ruins).

◦ Ch. 2: Non-linear exploration of a quest-dense urban hub (Calimport) and its outskirts.

◦ Ch. 3: Divergent racial paths and epic combat encounters!

How could Chapter Four be generalized?

March through the Underdark. It is full of dungeon crawling and battles against monsters with special powers.

What extra measures need to be taken in order to maintain game balance for epic builds up to fortieth level (since Swordflight will likely reach this cap)? 

I am still in the process of figuring that out. One key factor to keep in mind is that a player can make far better use of the powers of a high-level character than an AI, so for a monster to pose a serious threat to such a player it is generally necessary for it to significantly out-level the player. E.g., a first level mob is a real threat to a 1st level PC, but a 10th level mob is probably not going to be much of a threat to a 10th level PC, and the relative power of the PC is only going to get greater the higher his level (Of course I am talking about the effective level here, the AB, AC, HP, etc, not necessarily the nominal level which could conceivably be quite meaningless). It is also important to keep the loot under control, something most high level modules fail to do, so PCs cannot make themselves immune to almost everything that might hurt them.

Most people, including myself, are of the opinion that Swordflight design reached its apex in Chapter Two's Calimport. Is there any chance of another Athkatla-esque hub in the future, filled to the brim with non-linear questing and exploration, or will you continue to follow the pattern whereby each Chapter is unique, with its own independent look, feel and structure?

The latter. Sadly, I do not expect any of the remaining chapters to be as large or non-linear as Ch. 2.

What aspect of Swordflight are you most pleased with? Something you look at and think, "I really like the way that turned out! Heck, I surprised even myself!" It could be a quest, a puzzle, a character or a piece of dialogue. On the flip-side, with what aspect are you least satisfied? (Perhaps something that you intend to address in a future update.)

As a follow-up to the previous question, I am amazed that I was ever able to make something as massive and complex as Ch.2 all by myself, and in just several years. I am still kind of amazed that actually happened.

I was pleased and a bit surprised with how most of the dialogue and interactions with Zarala turned out. She was planned to be a hindrance and a nuisance for the player (and is) but it seemed to me she also really emerged as someone with a distinctive personality. I am never sure how the "non-mechanical" stuff like this (i.e., other than the combat, puzzles, special traps etc.) is going to come out before I actually sit down and write it.

By contrast, I am not altogether happy with the way Ch. 3 & 4 turned out. There is perhaps too much filler combat and not enough focus on the story in these chapters. There should be more of the latter in Ch. 5.

I am actually not sure I am the best person to answer this sort of question, since it often turns out that an author's favorite parts of his work are things most everyone else did not particularly like, while the favorites of the players/readers/etc. are things the author hardly gave a thought to. Muses are fond of playing jokes on authors in this way it seems. Though in yet another odd paradox, one of the best ways to produce stuff fans love is often to largely ignore what they call for and just concentrate on implementing one's own vision for the work.

After you have polished the fourth installment and tinkered with installments 1-3, I suppose you will get to work on the fifth. You once projected Swordflight to seven Chapters in total. Thus, is it likely the series won't be completed for five or six years or will you endeavor to speed up your releases?

I am always endeavoring to speed up releases, I just have not had much luck actually doing so. Ultimately it depends on how much time the rest of my life leaves me for this, which I cannot entirely control or predict.

I have altered my plans for the total number of chapters at various points, deciding to split up stuff that at one time was going in one chapter or vice versa. At one point I planned for there to be five chapters, and later decided I would need seven. At this time, I plan for there to be six chapters in total. It is possible I could complete each in a couple years or so, thus completing the whole series in several years, but it will probably take longer.

Do you have any further plans for making mods, etc., if and when Swordflight is completed?

I have no particular plans, but as I have found I rather enjoy game design I will probably try to continue doing it one way or another. Whether that means further NWN modules, modding for some other game, or attempting to make an indie game of my own, I do not know.

In my interview with Lukas Kristjanson (Lead Writer of Baldur's Gate), he mentioned BioWare's inability to employ a text entry system for player input of passcodes (in order to answer a riddle, for example). Since NWN overcomes the Infinity Engine's GUI restrictions, since you don't have to be concerned about translations, and since the text entry system has worked well in a couple of other modules, have you considered employing it in Swordflight?

Such systems have a lot of potential and it would not necessarily be a bad idea, but I have to admit I have hardly thought about it, mostly since the main uses for it (e.g., riddles) have only had minor and occasional relevance in Swordflight.

In regards to the Aurora toolset, what difficulties (if any) did you manage to overcome, and is there anything you would have liked to implement in Swordflight, but could not, due to the engine's limitations?

Getting a functioning pick-pocketing system in place, that would allow the player to actually get some use out of the skill, not force a reload on failure, and work with the ethical system in the series by means of appropriate alignment shifts was one of the hardest initial tasks I had to tackle. Naturally a lot of things were hard in Ch. 1 when I was first learning how to use the scripting system and other toolset features, and have generally gotten easier since.

I would have liked to have had an equivalent to the Tomes in Baldur's Gate that permanently increased Ability scores, but this was not possible in NWN (it is technically possible with NWNX, I believe, but it would not really have been practical to make use of that in this context).

For the most part, Aurora's limitations have not affected me too much, since I specifically planned the series with its major limitations in mind.

FPC in Baldur's Gate

Aurora has two major limitations that might be relevant to someone making a different sort of campaign. The first and probably most notorious is the lack of full party control which limits the ability to employ party-based tactics of the sort for which the IE games are known. This does not mean it is impossible to make challenging scenarios that require sophisticated tactics in NWN, as has sometimes been claimed (indeed, theoretically, a character with access to enough spells, special powers, usable items, etc., might actually have more tactical options than a party made up of characters each of whom has a strictly limited repertoire of powers), but it certainly imposes some limitations and affects the type and style of combat.

The second major limitation (of a quite different type and relevant to yet another style of game) is the lack of a fully-explorable, fully three dimensional game environment of the sort one finds in an Elder Scrolls game or something of that nature, where, e.g, you can come to a body of water and jump in and swim to the bottom, come to a cliff and start climbing up it, cast a levitation spell and fly up somewhere, etc. It is possible to simulate this sort of thing to some extent in NWN, but generally in kludgy and unsatisfactory ways.

How will the existence of Beamdog's Neverwinter Nights: Enhanced Edition affect the development of Swordflight? What counsel would you offer Beamdog, who are not known for maintaining game balance in their IE:EEs. For example, I have seen suggestions for four-class builds in NWN:EE. If employed non-optionally, how would this affect Swordflight and other adventure modules and campaigns?

Diamond Edition

Since I intend to continue the series in Diamond Edition (with EE being backwards compatible this should make the modules playable in either version, maximizing availability), EE should not have much direct effect (hopefully the number of EE-specific bugs that crop up will not be too numerous).

Given the crucial importance of backwards compatibility in NWN's case, I would advise Beamdog to be very careful and not change too much directly, while making their main focus be the reduction of the hard-codedness of the game to expand the options available to modders. For the most part, this fortunately appears to be what they are actually doing.

Introducing four-class builds as a default setting in the game (as opposed to an option builders can turn on if they want, which would obviously be a lot easier to justify) would likely have disastrous consequences on game balance in every existing module, and potentially also have negative consequences for other kinds of class-specific content. Even modules made after the change would be far harder to properly balance (NWN2's broken class balance comes to mind here – 4 class builds are not the only reason for that of course, but they do not help). 3rd Edition D&D is a system meant for multi-classing, and one has to remember that strong NWN players do not play classes like "Fighter," they play something like a Rogue/Fighter/Weapon Master or Bard/Fighter/Red Dragon Disciple. Such a change could make almost every existing build (and by extension any modules built around the currently available builds) obsolete. Naturally, I do not recommend it.

Lastly, I must know: Was Zarala firing a crossbow bolt into the PC during a cutscene a direct reference to Ian shooting you in the back in Fallout, or just a joke about dumb companion AI in general?

Perhaps this is disappointing, but I actually was not thinking about the Fallout companions at all (I think it is mainly an issue for noobs in Fallout – I soon learned to not give sub-machine guns to the companions and thereafter forgot about it). It was partly a joke about terrible companion AI in general, partly inspired by historical instances of military "friendly fire" incidents, and partly a warning to the player about what to expect from Zarala.

What I would like to know: Does Minsc think he is actually a Fallout companion? There is no telling what ideas various blows to the head might have given him, and his constant advice to go for the eyes would make a lot more sense in that game.

End of interview

Thanks to Rogueknight333 for taking the time to answer my questions, and for gifting cRPG fans with Swordflight, a campaign that made myself, and many others who have played it, rethink what constitutes good cRPG design. Haven't played Swordflight? Download it now.

Neverwinter Nights	Swordflight Chapter 3	Buried Ruins 2	Thieves' Guild	Mazar's Golems
Swordflight	Swordflight Alignment	Buried Ruins 3	Goblinoid Army	Precious Metals
Rogueknight Interview	Swordflight Walkthrough	Buried Ruins 4	Orc Army	Hammer & Anvil
Swordflight Chapter 1	Shifting Sands Inn	Calimport	Dragon Cult	Bloody Blades
Swordflight Chapter 2	Buried Ruins 1	Zataroth	Aranea Base	Zagash the Terrible