The Early Days of id Software

hello everybody Welcome to the talk ID software's early days as told by me um so welcome everybody I'm John Ram co-founder of ID software and I'm going to take you on a journey back to the beginning of ID software are you ready to be entertained everybody needs some entertainment today all right okay so I realized that some of what I'm about to say may sound insane but we were in our 20s and when we started it we didn't know that there were any limits so I grew up in a wonderful small town in northern California named Rockland the population was only 6,000 in the 70s I was massively addicted to spending loads of time in cold dark arcades playing all the games

there and I was got really good at playing them in 1979 before anyone really had a computer at their house including me I went to the local College when I was 11 and I started learning basic from the college students I went up just went up to them and I just asked them what the words in their listings were and I wrote them all down in paper and I started experimenting with them on the hp9000 Mainframe there I was mostly interested in just making and playing games to keep me at home my parents got an Apple 2 plus I was done going outside after that I spent all my time programming games on that computer a few years and 20 Apple 2 games later and I finally learned 6502 Assembly Language from this book and it was the language that all fast arcade games were written in back in the 80s

then I could really make 80s games like this except that they kind of look like this on a computer so not quite arcade games but home computer games which were on the Apple 2 and let's just say the Apple 2 was my personal home arcade as well as 1 million other Apple owners back then uh when I was a sophomore in high school I did some programming for the Air Force when I was 15 years old I can't tell you what was I was programming though it's classified uh After High School I kept making games and by 1987 I was finally working at one of my favorite companies in the industry which is origin systems my first job was porting 2400 ad from the Apple 2 to the Commodore 64 by this time in 1987 1988 uh IID made 74 games

and three previous startup companies Capital idea software on top inside out software uh where I ported M magic 2 uh from the Apple 2 to the commer 64 and ideas from the deep and I was 21 years old so then then I went to work at a company named soft disk at the start of 1989 I learned how to program a Dos PC there and I made a small game or a utility per month for a whole year then I created a game product called create uh Gamers Edge at soft disk and I had to hire a very small team of developers to make games so I hired John carac and Adrian carmac not related uh into the department for programming and art and then Tom Hall came in at night to help us out since he was already at soft disk and he loved making games so this was the first

time in uh 10 years that any of us had worked with one other person or more on any game after making them alone right for a decade each and it was incredible getting together finally on a team so while creating our first game together SLX John carac discovered the smooth scrolling truck on the PC so Tom and John stayed up until 5:00 a.m. and they made a a little demo called danger Dave and copyright infringement and so they put a desk on a disc on my desk the next day I came in I saw the disc I ran the demo and I watched the screen scroll smoothly pixel by pixel and it was a massive Eureka moment for me it was like a bolt of lightning hit me and I'll elaborate on why that was in a in a moment but it sare was born that instant on September 20th of 1990 so one thing led to another and we

spent a whole week putting together a demo of Super Mario 3 for Nintendo we did this after workor and Nintendo liked it but they decided not to publish it because they decided to only publish their games on their own NES platform back then which was a smart move so no problem we just use the tech for different game which was the commander King Trilogy so why would a sidescroller be a huge hit on PCS in 1990 well it was because no games on the PC had scrolled smoothly per pixel yeah the PC had been out since August of 1981 and in 9 years no one had figured out how to make smooth scrolling pixel by pixel like a Nintendo until the dangerous dve copyright infringement demo which led to Commander Keane does anyone remember the original Duke Nukem game all right uh it's scrolled by in

chunks of eight pixels like other games of that era and the reason why it scrolled with any speed is because I taught Todd rogal how to do it while he was coding Dark Ages and he made the the game that he made right before Duke Nukem so the commander King Trilogy provided the start of the company its software and we made these three games in three months from September 20th to December 14th in 1990 so commander King was a massive hit for us and it was so popular that people cosplayed as Keen for years at events even today and Commander Keen pioneered the creation of game engines so we designed the game as an engine that operated on different data for different games it was the only way to get the trilogy done so quickly in fact in 1999 1991 when we were working on keen4 we started licensing the engine for the first time so as the beginning of the modern engine engine licensing

Business Development on our games went very smoothly and quickly because we stuck to some core principles that are important even today so through this talk I'm going to highlight some of our core principles I'd like to highlight something else right now namely this photo has anyone seen this photo before so not in Europe um it's a picture of of uh John and I at the lakeh house in treport Louisiana where we started in software so the funny thing is that people have asked me for years what was in this picture so I analyzed it recently and here's what you see so this is John me and John in September early September of 1990 we were working on the Super Mario 3 demo that we planned on sending to Nintendo so we both worked on this huge Dungeons and Dragons table that John had we used to play D and D on the weekends and those sessions led to ideas for future games like doom and

Quake so this is John and me Tom Hall took the photo um we were working on the Super Mario 3 demo that we planned on sending the Nintendo and the computers were brought home from work on the weekends so this pick was taken on a Saturday or a Sunday on top of the monitor is one of those old Intel reflective astronaut plushies from a long time ago and on my left is my notepad which was a task list of bugs to fix and this is our high Lev task list of what had to get done to finish the demo there wasn't really anything online back then there was no online really uh this is Tom Hall's area where he was doing all the graphics for the demo he recorded gameplay on a VCR then he played it back and he paused the action so he could duplicate the ti tiles exactly in Deluxe paint 2 if anyone remembers that the TV set had a 13 Channel selector on it and it was connected by an RF modulator

so it was very old school so it sofware was formally founded on February 1st of 1991 and we made 12 games that year Shadow kns dangerous save in the Haunted Mansion rescue Rover hover tank tons of games we actually took two months per game but we made two games simultaneously so this was due to basically having 10 years of intense game development experience prior but it's also due to the first principal that we had which was no prototypes just make the game polish as you go and don't depend on polish happening later so always maintain constantly shippable code this is how we made so many games so quickly we had the whole game designed in our heads you know we just needed to quantify what needed to be done and then went about working on it until the game was finished there were no prototypes for our games so we just made them and remember we did have a small team of four people that could do this and large

teams definitely require more planning so time for a quick Story one day it rained really hard and uh Cross Lake in Louisiana Rose and it was flooding everywhere and I needed to get to work we were furiously working away on our games and I had to get back into it and I just took a shower and everything and went down the road in the car and then I saw this the whole Road was flooded well basically I waited through the the huge flood with water snakes in Louisiana all the way to the house and they took another shower and they could start coating and this was because we're all so excited to be making our own games you know 24/7 basically also note that during this time uh that of during this year that we made 12 games we also moved ID software from Louisiana to Wisconsin it's Madison Wisconsin and that takes a lot of time out of making games but we couldn't afford to have anything go wrong while making our games at such a pace and so

we created another principle that kept us developing quickly it's incredibly important that your game can always be run by your team so bulletproof your engine by providing defaults on load failure so if you have 100 people on your team trying to develop a game that will not run you're paying for a 100 people to sit around and wait for it to get fixed so we recognize the importance of keeping the game always playable and we decided to Bull prooof our engine by making all input into the game solid so game engines typically fail because they're trying to load bad corrupted or non-existent data so checking for this and providing a default for a failure case will keep the game running and it'll quickly show you what's missing so if you fail to to load a Sprite into the engine just show Bagel uh theme song is not working uh or loading up play something obviously wrong for the game like I don't know heavy metal or something uh missing a sound effect same thing just play something very obnoxious

so after 1991 it software's first stage of company development was complete and another important principle was in effect keep your code absolutely simple keep looking at your functions and figure out how you can simplify even further so we wrote all of the games that we made up to and including quake in plain C no C++ not saying that you shouldn't be using C++ so um so okay stage two was about to begin in August of 1991 we decided to move to Madison Wisconsin Tom Hall and I visited at that time in August and we found it to be really nice because Tom Hall used to live there while he was in college so we moved all four of us there and we continued working on our games only four months later we were found dead in the snow victims of Wisconsin's brutal Winters that we did not research so the moral of the story is do your

research we knew how to program an assembly language but not how to ask Tom Hall hey what are winters like up here so after 6 months of of cold hell we moved to Texas so new principle great tools help make great games spend as much time on tools as possible I wrote a tile editor in 1991 named Ted for tile editor uh t head was used for 33 shipped retail games uh several of which were 3D games even like hover tank catacombs 3D wolf and sign 3D Spear of Destiny rise of Triad Corridor 7 Etc so it was January 1992 we decided to go all 3D based on catacombs 3D's promise that we had we we were had just made it a couple months ago so it looked cool it just didn't play very cool so the wolf wenin 3D it took us 4 months of

development time to make Wolfenstein and launch the shareware version with one episode of you know one episode of levels it took two more months to finish all six episodes of the game plus the hint book and then the first month the game sold 4,000 copies all priced at $60 each so that was really pretty cool for us uh spirit Destiny took two months it's a prequel to Wolfenstein 3D and it was retail only so we got two games done that year soon after Tom Hall traveled to Kentucky to work for a couple months on Wolfenstein VR yes this was 1992 VR back in the days of Commander keing I had discovered a small three-person game company called raven software in Madison Wisconsin I called him up we went over we introduced ourselves and flashed forward 7 months later and we did a little work with them by modifying the

Wolfenstein 3D engine and Licensing it to them for a game called shadowcaster if anybody remembers this this game it was released in 93 shadowcaster Tech improvements were sloping floors lighting and fog this engine looked slightly better than Wolfenstein 3D but it just wasn't good enough for our next game so John carac he spent some months thinking about how more advanced the new engine should be for the game that we decided to call D based on the rapid development of our previous games we came up with another important principle we are our own best testing team and we should never allow anyone else to experience bugs or see the game Crash don't waste other people's time test thoroughly before checking in your code no throwing it over the fence for testers to find and put it in a bug put a bug in a database and then fix it later it's a wasteful cycle so after 1992 it software's second

stage of company development was complete along with another Principle as soon as you see a bug you fix it do not continue on if you don't fix your bugs your new code will be built on a buggy code base and ensure an unstable Foundation if you check in buggy codes some will be someone will be writing code based on your bad code and you can just imagine how much waste that creates okay so the ideas for Doom came from our DND Campaign which is full of demons we also love the movies Evil Dead and Aliens so Doom's design was just a mashup of these ideas and at the beginning of Doom's development we created a new core principle use a superior development system than your target to develop your game so before Doom we were making games for Doss while developing onos computers

and we knew that we could do better if we used more powerful computers and a more advanced operating system to develop our games so we developed doom on nextstep workstations so they're far superior to Dos Next Step eventually turned into OSX OS 10 this also meant that one of our core principles was upheld great tools help make great games we could make far better tools on next step so doomed and Quake Ed were two of the most important tools that we used they both really helped us create levels and test them very quickly you might not have known this but we had five people on our team creating Doom after Tom Hall left we hired Sandy Peterson and Dave Taylor which brought us up to six people we also did the Super Nintendo Port of Wolfenstein 3 3D right in the middle of Doom's development it took us

3 weeks to make this port because we had to learn the Super Nintendo Hardware so we uploaded the shareware version of Doom to the University of Wisconsin server on December 10th of 1993 the excitement for the game was unprecedented people were sitting on the server in upload directory creating files with sentences like when will we see doom and we got random calls in the office asking when it would be out so time for a quick story during the final day of Doom's creation we worked 30 hours that day so we had this we had the game running on all the computers in the office to ensure that the game was was really solid so on a couple computers the game froze the menu could be brought up but the game play like just it just stopped so John carac thought about what could possibly be happening in the game

and he figured out the solution without actually doing any debugging so he was correcting his solution and we finally uploaded it after this 5minute fix and more and more testing so at the beginning of 1994 I started working with Raven software and developed heretic using the Doom engine I wanted to see how an inventory system in a medieval version of Doom would play it turned out great I think do does anyone remember heretic okay it made it yeah we also made Doom 2 in 1994 over8 months and it was released October 10th in addition to this we did the Jaguar Port ourselves and again we were multitasking and making multiple games so we made two games in 1993 three games in 1994 1995 we started working on quake and we got up to nine developers there were four designers

three coders two artists and I was the only one to actually do both coding and design I wrote quaked I experimented with level design in full 3d again we started with a clean code base we had no code from Doom using Quake which was another one of our Co core principles of development up to that time write your code for this game only not for some future game you're going to be in new code later because you're going to be smarter also you're not tying yourself down to the limitations of your past code always try new things so Quakes engine was being developed by John carac and the rasterization was being developed by Michael abash John Cash worked on the network code and he went on to become the lead programmer of World of Warcraft time for another quick story so this relates back to our belief that developing in a superior operating

system can result in a better game so while making Quake we actually had a deal with cray supercomputers to buy a cray ymp for half price so our plan was to have our development team connected to this cray to bsp and light our maps at supercomputer speeds as well as crunch whatever kinds of new data that we thought that we might need for our next games engine so the computer room in quakes dm3 level was going to be full of cray computers as a reference to this new hardware that we're going to acquire unfortunately cray was bought by silicon graphics and before Quake was done and that deal fell apart so the computer room in Quake is is filled with the usual rectangular main frames instead of c-shaped cray supercomputers after publishing heretic I started working with Raven on hexen I wanted to see how an FPS would play with a hub level system and character classes on it hexon launched on October 30th of

1995 during the death match 95 event that was happening at Microsoft headquarters in Redmond so a month later I got Raven started on my next design which was hecome it would be the third game in the series heretic hexen and hecome hecome was never finished it was turned into hexen 2 because I left it uh during this time we noticed a small game company making some nice games like Raptor call the shadows and we brought them down from Illinois to make a game that we would publish they called themselves Rogue entertainment and about 14 months later they released Strife which used the Doom engine it was an FPS RPG and it was really fun and it showed that the combining genres could actually make a fun FPS has anyone played Strife here all right it's on Steam by the way um also during 1995 we made ultimate

Doom which was a re a retail version of the full version of Doom with an extra episode and we made the master levels of Doom that year as well so it softer was still nine developers in size and we released two games in 1995 while we were working on Quake so work continued on quake and 14 months after we started we released Q test on February 24th of 1996 for the world to test our first internet gameplay during the next 4 months we worked very hard to complete Quake we also released final Doom which was uh created by team TNT and the Cali brothers and we also released death Kings of the dark Citadel which was an additional set of levels for hexen so One important principle that helped us get Quake done faster was this one encapsulate functionality to ensure design consistency examples of this in

Quake would be like just the torches on the wall we could have had us level designers place a torch model then a fire model That animated then a torch sound entity all at the same location in a map but then if we needed to move things around something could have gotten left behind so it was far easier to just create a torch entity that we dropped in the game with all the functionality built in and spawned by the engine so also the water in the game needed sound effect entities all over the place to fully cover the water areas so you'd hear it everywhere so if the water got modified in the level moving all those entities around and deleting some it would have been a huge mess so it was easier to make the game just play water sound when the when the water was being rendered so it was a renderer level feature and it was out of the designer's hands and that ensured consistency and it saved memory and we did the same thing for the skies in Quake Sky

audio so I released Quake shareware on June 22nd 5:30 uh Central Time on the University of Wisconsin at Madison site so quick story while Michael abrash was programming the renderer he was interleaving CPU instructions with FPU instructions to calculate prospective correct texture mapping so sometimes while he was playing the game for one frame the game would show a completely different part of the level he isolated the only instruction where that result could actually happen and determined that it was impossible for it to be an invalid value so he had a friend from Intel come over and go through his analysis and his friend agreed with him and told him that there was a known error with the floating Point divide instruction on the Pentium and it was a hardware error so there was nothing that we could do

about it so we left it alone but this bug is known as the Pentium FD bug which you can look up online so Quake is the game that introduced the world to Mouse look to high-speed true 3D worlds texture mapped and internet multiplayer so Clans sprung up immediately as soon as the game came out and Esports and tournaments were starting all over the place Quake world launched 5 months later to smooth out the internet gameplay so making games was and still is our life we love it more than anything else as you can tell by our release of 28 games in 5 and 1/2 years by less than 10 people many other games were released that use our license technology over the years and here are some more core principles that we learned from doing all of this work try to code transparently tell your lead lead and peers exactly how you're

going to solve your current task and get feedback and advice don't treat game programming like each coder is a blackbox project could just totally go off the rails and and cause delays programming is a creative art form based in logic every programmer is different and will code differently don't waste time focusing on a rigid coding style it's the output that matters so we keep working to make great games it's in our blood as it's in your blood and it's why you're here at GDC so you want to be the best go be the best thank you for your time and I'm ready for any kind of questions you might have [Applause] [Music] thanks so yeah I don't know if someone's handing a mic around or if anyone has any questions about this stuff okay you got one all right who who wants to ask some questions first or does nobody want to ask a question oh got one right over

here hey great talk thank you uh in one of the first principles you showed uh you mentioned that don't prototype and today one of the main principles in our game production as we learn as we work we are prototyping what's your opinion on that well we we because of the time pressures we we had no time to do prototypes and we were a small company that only had four people at the time so we could actually not we didn't need a prototype cuz we knew from our 10 years of experience that um the games that we were making were small in scope they were very easy to visualize in our in our in our minds and that when we started it and we finished it say one or two months later it was what needed to be done in that amount of time so you know there may be 20 levels in the game and it might tend to end up being 10 levels you know it's like whatever would

fit in that time but all the gam play and most of the the characters and and Ai and all that stuff could totally get done in that in that space um but I think nowadays obviously with any really big project you're definitely going to be prototyping but if you're still like a lot of mobile games they might not need prototyping if you already know exactly how the game is going to play but if you don't know if you're developing your game kind of organically and you want to see where the game goes I don't know if I would call it prototyping I mean it's like if you start making your game and at some point you might discover something about it that changes what you're doing in the game but that's your game it's a prototype if you're throwing it all away and starting over but if you're still going forward I don't know if that's a prototype I think that's just the beginning of your game thank you all [Music] right yeah hi hi hi um I always felt

especially looking back on Doom and Quake that that the music in your games was always a big part of the games I for example I I often listen to the to the Doom soundtrack and um I was just wondering for example the the collaboration with Tren rner on Quake how how did that happen maybe you got another quick story that you could share yeah so the the story with how we got Trent resar involved in Quake was kind of funny um after Doom was it was it hit everybody Doom was was such a big game that um Hollywood started to get into they tried to get into games so they they knew like there had been an earlier Hollywood invasion of the game industry and they were rejected and they decided to go in in a different way so um we were contacted by Hollywood agents uh ICM and we had a person handling our our account there named Bill block and Bill block at that time had um a lot of other clients and one of which was was uh 9-in

Nails well 9in nails they played Doom all the time on the tour bus they had four computers all networked on the tour bus when they were done with their concerts they were on the tour bus death matching and they did it all the time I it's funny because I just spoke to someone like two weeks ago about that like they they were were interviewing Chris Verena and getting all this like oh we play Doom all the time stories and so bill um found out about that and he said hey they're clients of mine and we heard that they were clients and we said wow it would be really cool if they would do the soundtrack for the next game instead of having midi kind of metal soundtrack we could actually get some industrial real music in here so um so basically they agreed that they they agreed to to meet up with us and they came over to our office in mosquite you know Trent would pull up in a in a limo and I don't know how but there'd be a bunch of women outside the building waiting I you know don't tell them anybody that they're going there but they find out where he's going and so

they just wait outside while he's up there and we're we're talking about stuff you know about the games and and the kind of music that we're thinking about doing and uh and we hung out several times and just kind of kept in touch uh through email and American McGee would fight into New Orleans where where Trent lived and um and just basically kind of coordinate the Music Creation and it was kind of funny cuz Quake was delayed because of the music because if you involve like someone like Trent who's like out on tour all the time and has a ton of uh a ton of Rights around the music there were 12 signatures we had to get from inner scope to actually get that music released to us so it took us extra time and finally as soon as it was signed off then we could actually release the game so just watch out for that thank you he was nice

okay so you had a rule stating that you didn't want to re reuse code because you would um not get the results that you wanted uh at what level did you decide between using engine code and specific game code because you use Commander Keen engine code several times over at what level do you make that decision and how well the commander keing code let's say for Commander keing 1 through 3 it was the same engine so it ran the data that we created so we created levels and and characters and audio and all that and that was just data the engine worked on and the engine itself had if defs basically in it for each of the commander ke back then so if if this is Keen one then these actors would be you know would would their code would be uh you know would be running um and we would load these files um so there were if defs in the engine just to handle the specific sets of data that that or sets of code that were different for the

different we just didn't want the executable to be full of three games worth of specialized information so we if deff some of it but um but and and to keep things really small on a dis we had to take um like generic data files and turn them into object files and then link them into our executable and then compress the executable cuz we had to fit on a you know 140 or a three what was was it back then it was a 1.44 3 and 1/2 in disc um so we had to we we we did need to do that where we we didn't keep all of our code in one big executable but it was it was if Deft out so in the code it was there but not in the executable um but when what I was talking about was when we went between major jumps like Keen Keen 1 through three to Keen 4 through six those engines have no shared code in them or Wolfenstein or or Doom or Quake it's basically start with a new code base and

that was actually back in the'80s that was that was pretty typical like if you look at the Ultima games uh you know like L british's ultimate games every one of them was just brand new code just start writing a new game and maybe there's a faster way to render tiles on the screen or you know use code paging where you're you're overwriting code in memory because the game's too big Etc um so yeah we we didn't we didn't um reuse code ex unless it was something that didn't it was very small and it didn't matter like loading a file but usually it's so small that we probably just type it ourselves again the runner all right um um now obviously you were working on um por across you Mar brows free um during

the 80s stroke early '90s um why do you think Sid scrollers even though they were really popular during that era um even though they fallen out of favor now um why do you think as a pioner of the first P Sho person shooter genre why do you think this genre has gone from strength to strength to strength to strength and has no there are no signs of it like going down in terms of popularity you mean the shooter genre yeah the first person shooter genre that ultimate IID Pioneer back in the day uh well the the the great thing about the shooter genre is that it started to fragment its design across different styles so uh when we were making our games like we just made this this one type of game uh which I mean I'd say Wolfenstein and doom and Quake are very similar in that you go through levels and you just like kill monsters and get to the exit and it just looked better each time and and things acted differently and and sounded better all and looked better but but um you know as soon as other people started

to get into making shooters then we started to see people needing to be like more creative cuz they're not we're not the they're not the only game in town you know now there's a bunch of people making them so if you look at say the redstorm games when they started coming out with Ghost Recon Rainbow 6 and and Splinter Cell those were that was basically saying here's a whole tactical section of the shooter genre where you're creeping along um you're getting to do really long distance um sniper sniper rifle shots and and uh and and so Call of Duty where you have checkpoints and everything advances between checkpoints so people started to really try and innovate within the genre and now you have you know OverWatch you know TF2 and and uh and uh law Breakers in in in in this whole competitive way of playing Shooters and I know that that uh we have let's say MMO shooter Destiny so

everybody's taken this Viewpoint and this way of basically shooting stuff and applying it to a lot of different ways of getting through you know or or getting through objectives basically and wrapping some kind of narrative around it that's really different and sometimes you get something that's you know very different like Splinter Cell is is pretty different from those other games um but you're still you know you're still shooting in some parts um so I think it's just because uh there's so much competition people are trying to do something new with it um if you I has anyone played the The Hitman sniper game on mobile um because that's another really cool way of of making a shooter where you just have a vantage point somewhere on your mobile phone and you're just zooming into a scene and scrolling across a scene and tracking people and it's totally a first-person shooter but you don't move and it's just another way just another example of of how people are trying to innovate and push forward with this firstperson

shooter perspective but but how you play it is different and I know that the Innovation is just going to continue you know people are just going to continue trying to innovate with that Viewpoint and and what else can you do with it all right so yeah um so your story has been very inspirational to many many people in the industry and what do you think was unique to It software in those days that enabled the to happen and do you think it it can happen again to to teams these days and what would be your advice um yeah I mean I think it's it's uh super common you know really everybody can do the same thing and what it what it comes down to is people that match pretty well um having a pretty good amount of experience but really having focus and uh and determination to make your game and finish your game and do that over and over again so when um when Wolfenstein came out that was game

number 87 for me and doom was like game number 90 and we had made so many games and we had iterated and we have shipped over and over um that we were we knew how to do it and and we didn't have to really worry about how to finish a game or even how to start a game we were just worried about how to design a really good game and to create new technology and kind of make those things work together nowadays you don't really have to to worry so much about the tech and it really is all about you know design is law it's all about what is that design that you're doing that you're creating that no one has seen before because the tech is already there for you to use you know except for when you're in VR and AR and there's a lot of new stuff that that you're going to create there but basic traditionally on console and PC what is the design and can you focus and spend a lot of time on that you know Innovative unique design to make something people haven't seen before the engine's already out there if you look at Minecraft it's a perfect example of here's an amazing design that

one person created you know and and it was all about what you do in the simple world people you know it's easy to complexify a design that takes nothing to make something really complex it's hard to simplify everything and if you just sit there and spend time simplifying your your design taking taking away things from your UI and things that would just complicate what you're doing and try and make it contextual within the game then you're getting better and better at what you do and um and so I say everybody has the same chance to do it it's just it takes you know not really a team but it's helpful if you have a team and it just takes focus and determination and and I think for us it was us being in the same room and being collocated over the Internet it's a lot harder um given that you didn't make prototypes

and you made so many games were there any concepts that you had to walk away from or throw away um let me see if we um yeah we did so after we made our first Commander King games we were going to make a game called Quake the fight for justice and it was going to be an RPG and everything so we started working on it we worked on it for about 2 weeks and then we just decided we knew that Quake was supposed to be a really great game it's supposed to look great it's supposed to be a powerful game and that what we were making right after Commander Kean was not matching that idea of who Quake was um in our minds that you know from D and D so we basically said forget it we we're still making these simple kind of cartoony games let's just keep on getting better at it but we're going to put Quake off and so that was like early in 1991 so we waited 4 years before we actually got to the tech together and what what we knew we could deliver on the original you know promise or design

of quake in 1995 when we started working on Quake um other than that we uh we didn't really start on any game that we didn't finish it was we we talked about games and talked about ideas such as um when we finished catacomb 3D and we were planning on making uh the next set of Commander ke keing 7 through 9 um we decided that uh you know what let's not do the sides scrolling thing again I mean we can do maybe a keen someday in the future but maybe it'll be in 3D but let's start doing 3D stuff like we just made catacombs 3D which was really like one of the first texture mapped games even though it was in EG so um Tom Hall thought of this idea where um you know what if there's a a research facility and there's all you know mutants all over the place cuz they created these monsters and they're destroying all the scientists and you're going to go in and rescue them all and I just told him that's an awful idea I don't I don't want to play that game already and so um

I said why don't we just remake Wolfenstein Castle Wolfenstein and we all just immediately went oh my god let's do that so that original idea that that was just another idea that we had that we didn't do but that's like pretty much the only extent that we had back then of not making something uh it was we kind of knew what we were going to make and we just made them so there wasn't a let's start doom and then reboot or something like that we we just made made the games we just came up with the designs and made them yeah hello my question is you managed to make games so quickly so so very very quickly how did you do it and what happened afterwards because I think it became a little bit slower yeah it's kind of funny um a lot of new programmers don't fully understand what we had to do back then because there's so many apis nowadays that do everything for you so let let's pretend that in a world where there are no apis cuz that's what we lived in

there were no apis when we were making our stuff there was bios you know the thing that when you boot your PC and you see some of that text that you should never see ever again like that's your bios and that's the only thing that we could actually talk to through the simple interrupt 21 Assembly Language API that was it so putting pixels on the screen for us was we knew where in the computer's memory if we put a value in memory a DOT would show up and so that's how we put our graphics on the screen we actually had to write all the code in assembly to put the dots on the screen quickly so you'd see a shape in backgrounds and everything um there was no API to you know send a display list of of triangles to a a GPU or anything because it didn't even exist so while making Quake if you can just imagine a full 3d World everything that those video cards were doing we actually had to do that uh in our own code and we knew that we when we were developing those apis uh that that when we were

developing the code to rasterize Quake we were also working with Invidia on the pipeline that we wanted the cards to be using uh which are the apis that that turned into the the onard apis that direct X and open jail and stuff talk to um but we had to put every pixel on the screen ourselves we had to read the keyboard ourselves there was no nice functions to call to do stuff like that and that is just to get the tech working not to mention drawing characters you know uh in 2D so 2D key frame animation everyone knows how long that takes there was no tweening there was no 3D stuff so it took us a long it took us so it it sounds like it did it did take us a long time but we spent all of our hours working so um especially when we're in Madison Wisconsin why would we go outside there's no reason like to even go out we could order pizza and live inside of a one room so so that's what we did and um and so we basically slept for 8 hours and all the rest of the time

we worked for you know half a year while we were doing that which got you know several games done at that point um but really it was all about like we didn't have anything else going on there was no Twitter no Facebook no internet no nothing so nothing is popping up and bugging us so we could just sit there and code all day long listening to heavy metal and stuff so that's what we did there were no interruptions and it was all about time focus and determination okay um since she left um uh ID software uh Doom 3 came out and doom 2016 came out only recently uh Doom 3 um was criticized by some people for the fact they focused too much on Survival horror War Doom 2016 has been criticized um even though a lot of early Doom fans have um praised it for the fact that it's gone back to to Basics formula um a lot of people have also stated that it

hasn't taken into account the J innovations that have occurred since Doom 2 um if you were to be say like in an alternative Universe if you could have had a hand in the making of Doom 3 and 2016 how would you have um done Doom what what I have done for Doom three or Doom four um well I would have made a sequel I think I would have made a sequel to Doom 2 because Doom 3 and doom four were reboots of the original Doom in different in different ways so it would have been a sequel I can't say what that sequel would have been cuz I never thought about like making a sequel to Doom 2 if you look at um the games that we had made back then we made a game and sometimes we made a sequel to the game uh like Wolfenstein we did a sequel kind of prequel but then we did spear Destiny we did doom doom 2 quake and I started

designing Quake 2 before I left but it was always main shareware game retail game main Sherer game retail game invent new IP invent new IP invent new IP so we probably wouldn't have done that if I was there because we needed to create new franchises that was more interesting than living inside of a franchise for a long period of time since if it was an a game that was a service based game that was operational every day yes I could live inside of that for a long time but but we weren't making those back then and probably wouldn't I wouldn't have personally wanted to make a Doom 3 back then uh I would have that's why we made quake and so um after Quake two if I was still there we would have been done a totally different game so I guess the the answer is I wouldn't have done it all right six more minutes and excuse me for my a bit stupid question but I play in games today now and what's your favorite maybe

favorite title or general of video games now if it's not a secret oh say that again I couldn't couldn't understand that uh what's your favorite uh gen of the video games now today what's my favorite game or genre yes oh wow favorite genre well I kind of like fpss that's that's kind of obvious um I'm also love MMOs uh people you know people don't know this but I I um worked on an MMO for four years uh has anyone heard of gazillion entertainment so gazillion is an MMO company that I founded with two other people in 2005 and we worked on um I worked on an MMO that I designed for four years before we had to stop working on the game but during that time I played World of Warcraft every single day for 5 years 6 hours minimum 12 hours minimum on weekends so I have thousands of hours in wow that's so I love

MMOs um so we got five minutes yeah um so when you did your original games we didn't really have the internet as we know it today and we didn't have the feedback Cycles on for instance bugs as we have today we've also gotten more advanced in our engines so it's easier to create bugs how has that how did that how has that influenced uh game design and the creative process and do you think that's do you think it's a benefit or do you think it as a a bad thing um well yeah back then the only feedback cycles that I had back then was when I tried to get something published so I would have to send my games uh through the mail like paper mail with a disc in a envelope and then the editor at a magazine back when people for some reason typed games in from magazines in the computer um so I published a bunch of games like that and the editors would get the manuscript they get the dis and the big

huge and the big huge uh listing you know Assembly Language listing or whatever and I would I would they would see that they would review it and they would send back you know pages of little tweaks here and there uh you some of some of them were in the game most of them were in the text of the manuscript um and that was helpful because that actually they were usually really good suggestions and um and as we as we kind of went got into the age where people could just like get online and start posting feedback and we actually had beta testers who would send our discs in the mail to in the early 90s and um really good feedback cuz they would find some bugs in our games that we couldn't that that were not obviously crash bugs or anything just weird obscure little things and um and so that was helpful and then when we actually get to the point where we're on the internet uh people are just coming up with really good ideas and um you know there's there's you have you have to like obviously take

everything with a grain of salt and you can't read the comments and all that you have to as a game developer you trust yourself you trust your design um and and you just kind of have to believe that that you're making something that you like and uh but you do have to listen to criticism because sometimes you know people have a lot of really good stuff to say so I think nowadays it's it's it's gotten better games games are getting better um there's a lot of good feedback out there if you listen to it um and sometimes the players can be a really good part of the process of of defining where the games are going is as you can kind of see with the way Kickstarter works you know there's a lot of feedback going into Kickstarter campaigns and and people going forward I think we'll just do this one more what was that okay so uh yeah I'm just starting out

with programming well like two years and uh I'm starting older than you so this is kind of like a philosophical question more um how how did you when you started learning to program avoid burnouts or did you did you just program and you just had fun at it or there was no burnout no like I I couldn't wait to wait like in the summer times when I was going to school or when I wasn't going to school like get out of you know high school ends and and then it's like at 8:00 I could wake up and I could code until 2:00 or 3: in the morning I couldn't wait to do that I actually used to get mad that I had to go to the bathroom so there was no burnout because I was so excited that every time you know my the kind of programming that I like to do is I like to see results all the time I don't like to build a giant chunk of something that hopefully will be cool at some point and I'll see it I like to see stuff as as much as possible so um my iteration time is just usually a couple

lines of code before I look at it again I've run my code over and over like thousands and thousands of times I'll write a couple lines I want to see how it works so I know that what I wrote is not going to turn into this big huge problem later and um and so I'm always working on getting cool stuff up on the screen so I'm always excited the whole time I'm making something um so burnout is maybe you know you need to get outside and eat and and see something other than the computer screen um but anyway I really I don't burn out that much I think I'm the last question okay yeah that was the last question all right that was that was the last question I believe so um so thanks everybody for listening to the story um