GameDev.net - An Interview With Geoff Howland

Age: 24
Years programming: 12
Where he works: Lupine Games (Owner/Founder)
What he does: He's the Lone Wolf, baby
Where he is from: California (San Francisco area)
Cigarette of choice: Used to smoke Marlboro Reds but quit a year and half ago
Beer of choice: Doesn't drink anymore either, used to drink Guinness or Pete's Wicked Ale
Why he gave the vices up: "My wife, and just deciding to do things better. I didn't like hanging around drunk people, they piss me off most of the time."
What he listens to while coding: "a lot of Madonna's stuff, REM, INXS, a lot of classical music (mostly Tchaikovski and Rachmaninov), musical pop operas like Jesus Christ Super Star, etc. Classical is great because it does repeat in the same chords often, and there aren't any discernable lyrics to distract you."
First computer: C64
Why there's no picture: "I think game developers should be kept in the shadows"

You mentioned to me once that you used to write stuff for BBSs. Is that where you got your start?

No, I actually started before the BBSs on other projects (mostly RPGs), then I did some BBS programs, then I did some BBS door games.

For the record, what would you say you "do"?

What do I do? Ummm. I do design stuff for the gameplay, plot, some of the character work, I program and I try to talk to publishers. :)

Hopefully you'll get a good response from that latest publisher

Hope so, it started out well, I should hear back from them early this week. I'm not counting on anything though, its the nature of this kind of thing to take FOREVER.

What was your M.O.S.?

7222 Hawk Missile Operator. I got gypped, I wanted to be infantry

How long were you in the Marines?

4 years. Not full time though, I went in Reserver, so I only spent about 1 year in full time for training and stuff, which was good because I didn't like my MOS much. Its not particularly bad, it just wasn't my thing.

Well, blowing stuff up is neat, too...

My actual job was to stand in the rain next to radars. I think it was my sub-MOS. Heh. I liked shooting stuff and blowing things up like we did in infantry training, not setting up radars.

So, do you have time for much outside of coding?

Outside of coding? What's that?

We don't know, we were hoping YOU knew.

You have any reference books on it? Any Oreilly nutshell books? What's the animal?

I have one... "Fishing for Dummies" but that's about it.

Ahhh. I'll look for it next time I go to Borders.

Do you spend much time PLAYING games? If so, which ones?

I wish I had more time for that actually, I really only get to play my PSX lately, because I need my PC for compiler, so between compiles or reboots I play games on that. Mostly split between driving games like Test Drive 4 (haven't gotten 5 yet), NFS 3, or fighting games like Soul Edge. Just beat Metal Gear recently too (a couple of times). The games I have kept myself from playing but WANT to play really bad are Half Life, Grim Fandango and Rogue Squadron.

Do you avoid playing those games because you think they would take too much time?

Absolutely. Grim Fandango means I probably wouldn't touch my compiler for a week or however long it took to play through it. Half Life would probably take the same, and Rebel Squadron would mean I would take "breaks" that would end up going all night...

Have you played Commandos at all?

I played their demo.

My first impression of VT was that it was the same kind of game.

It's sorta like that, but only from the most basic approaches. Commands and VT share that they are wargames, concentrating on a few troops, but from there they separate.

I think the biggest difference between VT and Commandos is that Commandos is like a puzzle game. You have to move character X to position Y and wait and kill the guard, and then move character Z to another position and do something else. Once you know that pattern, you have solved the level and that's it. VT takes it from the totally opposite approach of not forcing you to learn any patterns or in fact, creating any real patterns of the enemies, besides basic patrolling areas, etc. Everything else is handled by the units' AI and it's done in a non-linear fashion. For instance, in the demo of the game, you have to destroy a tank. You don't have to kill all the units to do so though, you can just sneak in if you time it right. Or you can create a diversion and have the enemies looking for your guys, of you can kill some of the units silently and place the charge. There are a number of ways to go about doing anything, so you aren't limited to just what I thought of to solve the mission.

That's refreshing. My biggest gripe with Commandos is that it seemed like there was only one way to do it, which led to a lot of (very long) reloading when you didn't do things exactly right. That, and it seemed like you kept doing the same things over and over. I could not finish the game because it got so frustrating doing the same things over and over.

I want to try to make VT non-frustrating, as much as it can be at least. I think annoying your player by having things like the game ending as soon as one of your units dies is totally obnoxious and ruins any immersive feeling the player had with the game. It just totally removes him from being with his units, and also removes the responsibility of letting one of his units die. In VT if your unit dies, you can leave them dead and continue, getting new units who are just less experienced. The morale of all the other units will go down of course when you let this happen, but in some cases it may be a best case scenario. For example, you may have a surprise objective pop up in a mission and decide to try to do both the original objectives and the new one, which could be REALLY hard, so you make it through and do both, but you lose one of your men. Thinking you may not be able to do both again, because you got a little lucky, you decide to leave one of your men behind, and you pay the penalty of low morale and a new recruit for it, but it's your decision, instead of the game's.

What made you decide to do the lonewolf thing, and not apply for a job at a company?

I've always been really selfish about my game programming wants/desires. I didn't want to work for anyone else or do cool games for other people and then never see any of the rewards from it, so I always have tried to do it on my own, which until recently meant I didn't do much. Then last year I got everything together to really put a lot more time and effort into it, so I found an artist and started at it.

So about a year ago is when you started Veiled Threat?

VT was about 8 months ago, a year ago I started getting more acquainted with the different kinds of APIs and stuff and making some test games just to make sure I can do it. Before then I always tried making some MONUMENTAL game without doing anything smaller first, so I finally got it through my thick skull that you have to start small and work your way up, so I did a couple of smaller projects before finally creating VT, which turned monumental pretty soon afterwards (as all my projects tend to).

The WIBCI principle. Wouldn't It Be Cool If... and you keep adding things.

Right. Actually, the game doesn't feature creep so much as the whole idea is very non-linear which just creates a lot of work.

What, IYO, are the relative strengths/weaknesses of an isometric interface from both the POVs of player and developer?

I think there are some definite advantages to the player in comparison with a 3D rotatable view, which would be that the graphics can be more detailed, as everything is prerendered, the computers can be less powerful, so P100s can run the game fine, and it's not as confusing to try to rotate the view, which can take time away from playing the game, and for game players who aren't as used to 3D controls, rotation is a bitch.

I feel like, too often, a game is made 3D, even though doing so adds nothing to the game, and may even detract from it

Absolutely. It's a marketing feature now. In my way you can ALWAYS see all of your units, no matter where they are (unless they are inside the same building, but on different floors, which isn't as often as anything else) so you don't have to worry about changing views, you just see everything.

So, buildings have multiple levels?

Yeah, we have 3 story buildings (and probably more later), where you can enter on the floor, go up to the top floor, go onto the roof, sniper out the windows, etc.

VT is real-time, right?

Yes, although I'm thinking of adding a pause option for more detailed commands; I'm not sure about that though. I want it to be real time but I don't want people who want more to be restricted either. Currently it's real time only though.

If you overdo the real-time aspect, you can wind up with something that behaves more like a multi-homed WalMart $1.99 novel than a RTS. On the other hand, if it is done successfully, it can add depth.

There won't be too many options. I'm trying to keep things detailed, but simple. I want everything to be instantly accessible to people who know the system and not confusing, but also really easy to step up to. There are 4 units, so far I've found it really easy to keep track of them all. It's really very easy to pop around to the units, and you are usually using them all for a common purpose, like setting up a unit or two as snipers or spotters to see places your unit penetrating the base can't see or something.

I was really into X-COM: Apocalypse, but I could never play it in real-time since so many units would act insanely. Will your units perform intelligent actions if you are not controlling them directly?

Yes, they will stay out of sight, silently kill enemies if they approach (and hide their bodies), or follow any orders you leave for them.

As a gamer, I find it extremely frustrating to fork over $50 for a game and then have to spend 10 hours in the manual before I know how to play a simple scenario.

My inspiration on that aspect is actually always Street Fighter 2, I always try to think in terms of "What would a person who just walked up off the street be able to do with this?" then I move in steps higher "What would someone who knows Windows be able to do with this?" then "What would someone who knows the regular RTS games.." I want to try to make it so that its totally intuitive. The first job I give to new testers is to play it without reading the instructions and tell me everything they could and couldn't do, what was confusing, what they thought didn't make sense etc. Then read the short instructions and do it again with a new review, to try to eliminate those large learning curves.

How long would you say you work on VT a week?

Probably about 50 hours to 60 hours a week. Sometimes more

How is your wife with all the time you put into VT?

She's amazing. She's totally supportive of all of this and everything I want to do and she makes it really easy for me to spend the time I need to on it. She doesn't play games much though, except driving games sometimes. I've got her egged on a little about those as she doesn't like things she can't beat me at in at least one area. (She can kick my ass in every other way... heh)

Are you working too, or is this it for you now?

Yeah, I have a day job as a sys admin.

Geoff, you really seem to know your stuff when it comes to game design. VT has obviously had a lot of thought put into it. We have all read your articles on the subject, and think they are excellent. Where did you get your insight?

Thanks. I have been thinking about game design and working on game designs for the 12 years I've been programming, so I guess that's it. I just played every game I could get my hands on (and still do, all their demos at least), and weighed their good and bad points and eventually I think it gave me a "taste" for how to weight good and bad things.

Do you find yourself picking apart every game you play?

Yeah, I definitely do that. I now have a cycle of comments internally on games I play. First I try to just get the "feel" of them, and ignore the actual technical details. Ill play it for a while, like that, then I will start technically deciding how they did things. Sometimes it doesn't take more than 20 seconds to know just about everything about a game though.

As Lupine continues to grow, do you see yourself moving into more of a Producer/Designer type of role, and move away from coding, or do you always want to program?

I think I will probably always do some programming, as I like it in a lot of ways, but I am definitely more interested in design in the long term. I don't want to become some "producer" that just deals with things in the highest levels, I love creating the algorithms for the code and deciding how things can really work well, maybe it will be basically AI programming instead of other things, but I will probably move away from graphics or other non AI/gameplay things later on, just because I cant do everything myself forever... (not enough time in the day unfortunately, I have a real interest in 3D and network programming and graphics in general, but I wouldn't have time to get my projects done if I indulged in them).

How far along is VT at this point?

It's hard to say, but I think in an engine only sense we have gotten pretty far, perhaps about 75% on the basic engine, only about 10% on the AI (or less compared to my long term goals), content is probably the lowest on the scale though, as we only have 3 maps. But all of these things are sorta subjective too, once we have the first tileset done, its easier to create derivitive ones, like on the same type of terrain. Same with my AI code, I have had so little time for it because I've been stuck getting other things that are more visible done first, so it will develop quickly soon.

Once you get some kind of commitment from a publisher, do you intend to bring more people on board?

I'm thinking about that. I know I'm going to hire a writer to do my dialogue and flush out the plot for me into more accessible for the main stream readers, I'm not sure about programmers or artists though. I have to see how much help I need before Ill know, I may just contract some peoples time to do specific things if I don't need a lot more help (which I don't think I do now), for this project it might take longer to bring someone up to speed and guide them (since I don't have any managed to handle the people managing stuff) then to just do everything myself...

Do you have a target date for finishing VT?

There are a couple of target dates and it depends when I get funded, and how much they give me. If I get funded very soon, I could finish this in Sep-Oct. If they give me more money than I'm asking for, I may be able to finish it a month sooner, but I probably wouldn't want to and I'd just make it better. The time from now that I am not getting funded is basically pushing back this date as I can't work nearly as much (or as well) as I would be able to full time.

So would you have any tips for those amateur programmers inspired by you to follow with what you have done?

Sure. First, you have to know if you REALLY want to do it my way. My way is REALLY hard and there are a lot of things that are easier if you start up with a company and make industry contacts and then move on your own a few years later. If you do decide you want to lone wolf it, then you have got to be REALLY consistent with producing things. My goal is to see something new every day I work on the program, without visibly moving forward you lose motivation and there is no one paying you or on your back to finish the game, so without constant motivation your project just dies immediately...

Well, shifting over to the technical side a bit. The VT graphics paradigm is interesting. An isometric engine that uses square tiles. How, exactly, does that work...and why?

I think a lot of games use this actually. (Like Seven Kingdoms and AoE). Basically, the graphics are all modeled in 3DS and then rotated to the somewhat Isometric perspective (its not true 1-1-1 ratio, because that doesn't look as good), and then my artist cuts up the tiles and makes them tile together so there aren't seams. I was originally thinking (and built) a true isometric tile viewer, but its only real advantage is for drawing walls, like a Diablo mazish type environment and its REALLY hard for the artist to tile them together correctly, because they are diamond shaped...

Ah. Has artwork been a major problem for you?

Not at all, my artist has been great, I just tell him what I want, send a few pictures maybe and then wait till I get a prototype, make some suggestions and then I get the final product, its been really smooth. Interestingly enough, we've never met or talked on the phone, we do almost everything through e-mail.

A lot of startup developers have difficulty getting art or an artist. How did you begin your association with this guy? How did you know you could trust him?

I found him from just emailing a lot of artists on the internet after going to their site, I was talking to a few people and I just seemed to like his attitude better than the others, which turned out to be very lucky for me. I asked him his price and it was something I could manage, so that's how it started. As for trust, I didn't know I could trust him originally, but you can never tell this, so I just went on faith and it worked out well. Since we were doing a contract situation, and not an employee-employer situation it made it a lot easier of course if something had gone wrong, it was just work for hire for no determined period of time.

Reading through your diary for this week, you seem to have been having some serious troubles with DirectDraw. Was it just bad luck and a couple screwed up drivers, or is there a larger issue with the API?

The former. I never bothered to learn that much about DD or DX, as I really don't care about it at all as long as my stuff blits. I think I have everything worked out now, mostly thanks to a lot of feedback from my diary readers. I think a lot of people have problems with it, but I don't know how justified they are in it.

What ended up being the problem?

The problem was a couple of things. First I have a messed up video driver on my machine that wouldn't switch from 16-bit to 8-bit without freezing, then I think I may have had something not set up correctly, but I can't really tell because I've re-written it so much. Finally, I just assumed the pitch would always be constant, but I've learned it isn't.

Do you perceive many people as having serious quips with DirectX?

Basically, its an API like any API, and they all have their quirks and annoying things, people bash Microsoft for any reason, personally, I don't care, they are going to end up making me money by providing a standard format to work in. Actually, a pet peeve I have is people who bash MS for no reason. I don't care about MS at all, I have no like for them and I've had plenty of problems with their stuff (doing administration work on Win9x and NT systems and programming), but I don't think people can really expect it would be any different if some other major company was in the same position...

Being a certified *nix dude, I'm sure you have some interest in multi-player solutions.

Like ideas on how to make it better for multiplayer? Send small packets. :) I'm basically going to do it this way: Create 2 engine states, one for the network and one for the user. The network is the only true state and the user is just so things move smoothly on the screen. Update in small *datagrams* trying to keep minimal information, like unit orders and time so that the network-engine can update everything quickly and then re-synch the players engine locally.

Some people would contend that multi-player support in RTS's is a bit easier than other genres. Would you agree based on your experience with VT?

I haven't done the network code for VT yet, so I'll have to disagree on my non-experience but logical assumptions instead. While its true that RTS games don't have to update as quickly and their units are usually going to continue an order for a longer period of time, there are often MANY MANY more units in an RTS game than say a FPS. How many units could be in an FPS game? Max 8? 32? In an RTS you can have 500 units running around, and while most likely they will be grouped together so they can be included in an umbrella order, they might be split up into smaller groups as well (in VT they will be, but there will be less then 500 units). However, I could still end up with 60-80 different groups with orders, which out weighs the 8 to 32 max units in a FPS, luckily I wont have to update as much though... So it's not so clear cut as one being easier than the other, they both have problems. In a heated battle those 60-80 groups will change orders very quickly as their targets change.

You have certainly demonstrated your design aptitude. Now what are the problems that you face? For instance, in VT...what do you find to be the biggest hampering factor in your design processes?

Time is the biggest problem I have really, everything else can be done in time. My next task however is to go about creating a new AI state machine for switching to different modes of attack and for learning about different situations. The first effort of it is to detect and find snipers or hidden grenade lobbers, as currently the units can only attack people they immediately see and won't hunt very far for people they can't (as they don't have a memory of where they look, which is what the AI state machine is supposed to provide).

So, have you thought at all about what's next after VT?

Oh yeah, I have a ton of games I have already prepared for doing after VT. If VT does well I will probably have another group do a sequel to it, and maybe assist, but I have a lot of other things I want to do as well. There is a sort of racing game that I've wanted to play for like 15 years and has never been done, so I'm REALLY looking forward to doing that soon. Should be cool.

(Myopic Rhino Games is open to contract work for the sequel) ; )

Heh. That will probably go up to a publisher, I doubt Ill have a say on it, hopefully it will happen though, or I may do it myself I'm not sure. I am pretty cloudy on how much actual control Ill have on projects after this one, since I will then be tied more to people who are supplying the money.

Hopefully, there are still publishers out there that believe in creative license.

I know there are, you just have to gain their respect first, then they basically let those developers do anything, as we see too often. I think there is a lot of recklessness that goes on in this industry once people get the respect they need to do big projects, its unfortunate because it normally ends up hurting those developers and the industry in general as funding people look at developers like they are all irresponsible, whether the assessment is originally accurate or note. It's really hard for people who aren't yet known in this industry mostly because there have been so many failed projects for various reasons. While this is pretty natural I think its important to remember the cause of it. All businesses fail and software companies have the additional hardship of not only being businesses but also not having a product before they spend a lot of time any money, compared to a retail store that can get products and just start selling. Hopefully Ill be able to use my limited business sense to help me and the other people who work for Lupine Games to make it and not fold like so many other companies end up doing...

Well, Geoff, I know that a lot of people are keeping an eye on Lupine Games, including myself. I think the industry would benefit from having more lone wolves making it. I don't know if the days when the indies controlled the gaming industry will ever return, but the industry needs a breath of fresh air.

I think that the Internet may just change the possibility of that. The only problem now is money. You can make a really good team over the Internet (of course its hard to find dedicated people), but in a few years, the retail model may really change to make Internet sales much better, for boxed games and downloadable ones as well. Also with the abundance of cheap game dev tools that are professional quality, which will start happening more and more as well, (or near pro quality) you can get a lot more amateur developers too. The only thing is, the amateur developers have to understand that their strength is not in competing head to head with the big developers, but in working in their own interests and hitting niches that the developers cant/wont try to get.

Very true. Some of the more memorable games I have played have been from amateurs, at least over the past couple of years. Dink Smallwood and the games from Aliensoft spring to mind.

I think as time goes on, it will take less skill to create games, which will mean that there are a lot of crappy games release with no effort spent (or too little), but at the same time it will allow people with more of an artistic bent to make games where they wouldn't be able to before,and distribute on the internet, where they basically have the same status as almost anyone else.

Fun is important. Polygon count is not important.

I totally agree

Many amateurs feel pressured to make their games live up to the industry specs rather than to make the industry specs work for their games. How do you get around that problem, Geoff?

For me, I go for what will give the experience I want to deliver to the player. I use a totally Top-Down approach to game making. I first have a vision of what I want the game to be, then I cut up all the compartments needed to make that game. This totally eliminates the feature creep or need to the latest marketing gizmos for my mindset, as its not part of my vision of the experience, so I just ignore it. For other people though, especially consumers. They need to be able to get explanations to what things really are and probably see some comparisons on what things are "popular" and what things are good. For instance, who cuts up on Blizzard for not making 3D games? I haven't heard ONE complaint (although I'm sure some nutcase somewhere has made one) about Diablo II not being 3D. The graphics would drop, the hardware requirements would raise and the development time and bugs would increase as well. Having things like that pointed out to players is probably the best start.

I find the top-down philosophy that you mentioned to be very different. It sounds like it works for the purposes of an amateur game that tends away from the industry... but how does it interoperate with the publishers?

It works well in any circumstance. You just have to understand your medium. If you are trying to provide an experience that is good for the player as well as the publisher,then you expand your medium to do so. In VT's case, the medium was myself and the players. I knew I couldn't make a 3D engine AND the game, so the 3D engine was ruled out immediately. In the case of a publisher (and not worrying about the funds/resources for the 3D engine) a 3D engine is normally required, just because everyone has it, its the latest feature mania. (Even though people are finally starting to occasionally say "So what? Its 3d! Everything is 3D!"

So people are beginning to look a LITTLE more at the game, but not much so).

It's about time.

It wont happen ever completely believe me, eye candy is too important. Its important in the surface/shallow way but its also important to the fun faction. Its FUN to see cool stuff happen ,or make it happen. Its the whole experience that needs to be tended too, not just gameplay.

Speaking of the public and software... This seems like a perfect place to bring up another major issue in the game industry. The dreaded...you've spoken out against piracy on many occasions. I've learned a lot from your discussions with various pirates and from your debunking of their attempting debunkings... but on a more personal level... How do you forsee piracy affecting Lupine, or other amateurs?

It will fuck us up like it does everyone else, but more so as we are smaller and can take less of a salary cut to continue operations or expand to "normal" developer studio size. Piracy is really bad, but it will happen, I don't lose sleep over it. I hold people that do it in contempt, especially ones that otherwise say they care about games as they are actively destroying them by stealing the products... There will always be PC games but there is a growing tendency for stores to stop carrying them because they are not selling NEARLY as well as console games, and this has a lot to do with piracy. There are a lot of aspects to this and none of them can be simply tied up into something that can be discussed in an article or interview completely, but there are a LOT of things that could go VERY badly for the PC games market if piracy continues strong or grows. Hopefully we will have some more solutions to fight it soon, like the Disc Safe or whatever the name of that encoding format it, but better than that

Do you think that piracy is a bigger problem for the PC platform than for consoles? I mean, there is a big deal about emulation.

Oh yeah, most definitely. Piracy in its hard core form will ALWAYS happen. But just by definition hard core is MUCH smaller than the masses, which is where it really hurts in the money, especially since some pirates NEVER buy games, but casual game copiers often do. Console games, like PSX ones actually require a modded system, which means they have to fuck I don't remember how to spell that, haha) damn, they have to attach the chip to the board or buy a system that is chipped. Copying the CDs for PSX is NOT simple for someone without instructions and isn't straightforward. Cartridges are obviously much harder to copy too. Just this difficulty to copy the games eliminates like 98% (guestimation) of the would-be piraters, since its too much trouble...

Well, good luck. It is looking very promising and you are an extremely talented developer. I only hope that you prove something about the industry or the way it can be to all of us.

Thanks, I'm trying.

Rhino, Kevin, Nick, Jered, and TANSTAAFL conducted the interview on 2/21/99.