Classic Postmortem: GoldenEye 007 | 2012 | Europe Talk | Mar

Transcript

00:02 [Music]

00:06 hello and welcome to the gold night

00:09 double-o seven post mortem first I'd

00:15 like to present my credentials

00:17 I started making games when I was 12

00:21 on the abit BBC microcomputer this was

00:27 as a hobbyist and I would write in two

00:30 magazines with games and they would

00:32 print them for people to type in which

00:35 is fairly incredible today but a good

00:38 way to learn programming after

00:41 university I was lucky enough to get a 00:43

job at rare as a first place as a

00:48 programmer on killer instinct this was a 00:52

very hectic short by my standards

00:57 project of one year we had to bring an 01:02

operating system and a new computer

01:05 essentially was designed inside rare by

01:08 Chris Stamper and peacocks from the

01:13 ground up it was a thrilling project

01:16 after this I was twiddling my thumbs in

01:20 the stable block where the killer

01:23 instinct team was working wondering what

01:26 I'd be doing next but my most important

01:34 credential is this video that I made

01:38 when I was a child

01:41 [Music]

02:04 [Applause]

02:06 thank you

02:10 I'm full credit where credit's chew that

02:13 was a team effort

02:14 with me and my two brothers my youngest

02:18 was actually the star I never felt most

02:22 comfortable in front of the camera I was

02:25 on piano and my middle brother was

02:31 pulling down the gel so you know it's an

02:35 example of how you need to pull together

02:38 several technical effects to achieve the

02:41 design result now the beginning of

02:47 Goldeneye is a bit of a strange story as

02:53 I understand it rare was approached by

02:55 Nintendo and asked if they would be

02:57 interested in making a game for the

03:00 upcoming Bond film which didn't have a

03:03 title at that time only a number and I

03:08 assume there was some miscommunication

03:12 or at least it was a gray area

03:17 at that time there are essentially two

03:20 large teams that rare one I'd finished

03:24 Donkey Kong Country recently about 18

03:28 people and the other had finished killer

03:31 instinct coin-op that was about 10 maybe

03:34 11 people there was a little bit of a

03:37 competitive spirit between these two

03:38 teams but Tim Stamper who was one of the

03:42 directors of the company was very much

03:43 on me Donkey Kong Country team and that

03:47 was his brainchild that game and he went

03:52 down to he went down to leave Stern

03:56 where the film was going to be short for

03:58 a press junket they were promoting the

04:01 new film all the big stars were there I

04:04 didn't know anything about this at the

04:06 time his a his a scan from some of the

04:12 subsequent publicity

04:14 from that visit and here is a picture of

04:17 very rare I have to say photograph of

04:21 Tim Stamper who also schemed this button

04:25 on the left here with the art director

04:29 of gold my Andrew Ackland snow gold

04:32 neither movie this is and it says I

04:35 don't know if you can see this Tim

04:39 Stamper who will design the Super NES

04:41 game here's a photograph inside rare

04:47 with the very large supercomputer called

04:50 challenger in the end it was 20

04:53 processors and it was awesome in that

04:57 could do a great mouth rendering but not

05:01 so good in that I had to reboot it every

05:03 young every day or sometimes even three

05:07 times a day so this is kind of a just

05:12 kind of a funny of start Tim Stamper

05:15 decided not to make the film

05:17 Greg males on Donkey Kong Country who's

05:22 I guess you would say lead designer on

05:24 Donkey Kong Country wasn't that

05:26 impressed so they decided to pass I

05:29 heard about this as a rumor and I said

05:34 to Tim Stamper that this sounds cool

05:37 like I'd like to make this game and he

05:41 said ok and that was that was pretty

05:44 much my pitch

05:47 I don't really know and I can't really

05:53 explain why they had trust in me after

05:57 simply being second programmer on killer

06:00 instinct but I'm delighted and honored

06:05 that they did at that time I was

06:09 beginning to be involved in the hiring

06:12 and I hired my first ever interview II

06:15 was Mark Edmonds this was January 1995

06:22 and he's someone you don't hear too much

06:26 about from the Gold 9 team I don't know

06:33 if you ever have noticed that weird

06:36 thing that happens on one of the screens

06:38 which is a very tall person walking like

06:42 this you can just about make it out

06:46 that's the only presence of Mark Evans

06:50 in the game most of the team members

06:53 were in the game myself included were in

06:56 the game as a face on one of the

06:58 characters that would be randomly

07:01 generated all that was scripted but mark

07:04 might preferred not to be so in the in

07:08 the course of putting together my talk I

07:11 asked all of the Goldeneye team if they

07:16 had anything to say about the making of

07:20 Goldeneye I had great pleasure I'm

07:23 talking to each of them mark I asked the

07:28 first question I asked him was what was

07:31 it like working at rare and he was a

07:35 very quiet humble guy and he said this

07:39 in the nicest possible way he said it

07:41 was work lunch work suffer play games

07:45 for a bit work sleep repeat and you know

07:50 for mark and myself it was a two-year

07:53 eight-month project and about a thousand

07:58 days

07:59 it was it was a long project in the

08:03 beginning we were probably working

08:05 we were probably working ten hours a day

08:07 but by the end I'm far in excess of that

08:13 there was no such thing as an n64 the

08:18 name before was ultra 64 and the name

08:22 for that was project reality but circa

08:25 the beginning of this project it was

08:27 called the ultra 64 and the closest

08:30 thing to it was this Onix Silicon

08:35 Graphics Onix which I think the list

08:37 price was about a hundred and fifty

08:39 thousand dollars this is a no you can

08:43 see it's a very large box it wouldn't

08:45 fit under every desk and it emitted huge

08:50 amounts of heat for graphics cards and

08:52 it was not that reliable frankly like

08:55 all the SGI machines if you pay that

08:58 much money for something you know it's

09:00 it's not going to be all that reliable

09:02 and this was marks this was marks

09:05 computer and he set about at the

09:09 beginning working on the pipeline for

09:11 the game trying to get a character on

09:14 the screen trying to make do with what I

09:18 mean we knew it wasn't going to be able

09:19 to do very many polygons and we wanted

09:22 lots of characters on screen so trying

09:24 to make a character four-sided arms guru

09:29 shading that was the first piece of

09:32 graphics I ever saw on you know gray

09:34 shaded no texture no artist one of the

09:42 major themes I want to pull out from

09:45 this that I was thinking about a lot

09:46 when I was putting this together was to

09:52 what extent does it make sense to talk

09:56 about a game or making a game as a as an

10:01 endeavor led by one person I find it's a

10:06 very interesting question I'm in love

10:08 with movies

10:10 as well as games obviously but I love

10:13 movies and I consider myself a film buff

10:15 and from from the French critical theory

10:22 around the time of Truffaut etc they had

10:25 this idea of auteur theory which is the

10:29 idea that there was one author of the

10:32 work and that was Francois Truffaut or

10:35 whatever now this you know we looked

10:39 films perhaps too often films were

10:42 looking to gain to our films were

10:43 looking to books and you know this was

10:46 the idea that had an author wrote a book

10:50 and no toy is my way of saying well you

10:54 know myth maybe it's not like that at

10:56 all maybe it's a totally collaborative

11:00 process well when it works well and

11:03 there's no one Authority and there's no

11:08 one author so this is one of the things

11:10 I want to pull out recently I found some

11:15 paper from the Goldeneye project I

11:18 originally thought I had nothing but I'm

11:20 going to show you for the first time

11:22 about ten pages of very very early

11:27 documents that I put together which all

11:31 materials that I have this is my

11:35 jottings the date on this can't be right

11:38 it says 94 it must be 95 February 95 and

11:46 I was wondering to myself the two I

11:50 think Tim very important questions for

11:52 this game what's the appropriate level

11:55 of humor and what's the appropriate

11:57 level of violence this is a list of

12:01 questions mainly which is a good start

12:06 to any project I suppose

12:08 can we get footage can we use the actual

12:10 music come use a title sequence can we -

12:12 likenesses can we use material from

12:16 other Bond films

12:18 can we get photos of props costumes etc

12:20 a bit of a list so I was wondering all

12:22 this and actually in a communication

12:24 I made this clear turn in Tendo and they

12:26 were good enough to send back an

12:29 extremely detailed document explaining

12:31 everything we could use which was an

12:34 astonishing amount we're very lucky to

12:37 have such a broad license so these are

12:44 my thoughts at the same time in my mind

12:48 was virtual cop and doom on the ultra 64

12:50 as it was called no gun I'd also been

12:54 playing a lot of time crisis and

12:56 together with Mark and you know I could

12:59 see the value of having a gun in your

13:01 hand but there was no gun planned for

13:05 the n64 at this time we had no control

13:08 or no idea what the control would be

13:10 like at all so there was no there was no

13:16 certainty but there was a rumor that

13:20 there was going to be a analog stick and

13:24 that was all I knew so I was pondering

13:28 with the analog joypad be okay is it

13:32 okay to shoot lots of people I don't

13:34 think I actually put that question to

13:36 management or to Nintendo if I had they

13:40 probably said no it was very much in my

13:43 mind that bond wasn't I mean you know he

13:48 does kill people he does have a license

13:50 to kill but it's not a gorefest

13:53 it's not an incredibly violent game it's

13:59 not an incredibly violent universe so

14:01 you know this is very much in my mind

14:02 talking here about Street Fighter is

14:05 less violent than Virtua cop which is

14:08 less violent than doom which is less

14:10 violent than Pulp Fiction is an

14:14 adventure aspect a good idea gets the

14:16 camera route fixed

14:18 model destructions crucial character

14:21 motions we fixed can we use the likeness

14:23 of Pierce Brosnan maybe to play a

14:25 corporate ear for bigger idea legal can

14:30 we use the Walther PPK and down the

14:33 bottom I'm just starting to talk about

14:34 the schedule is it one programmer 2

14:36 artist

14:37 to two other artists are you know that's

14:40 very much think of a small member of

14:43 program as a large number of artists

14:45 controls aim shoot dive left-right

14:47 advanced retreat you know these these

14:50 are all ideas and then Tim asked me to

14:52 produce a a formal document so that I

14:57 could show the management of rare what I

15:00 was thinking

15:00 so I produced this this is April 1995

15:05 couple of months later slightly drawing

15:09 back my scope focusing more on virtual

15:12 cop which is obviously a simpler game to

15:16 make large number of character types

15:18 more moves and virtual cop for each

15:21 character this was all comparing it to

15:23 virtual pot more intelligent characters

15:26 this is where it starts to diverge so

15:29 this is suggesting that we spend much

15:34 more effort on the artificial

15:36 intelligence all characters will try to

15:40 survive but in addition they might be

15:43 trying to reach an alarm switch rescue

15:45 their comrade obey an order play dead

15:48 hide under a table or reach an exit

15:51 wouldn't it be awesome if a character

15:53 tried to hide under a table in the first

15:55 person shoes I'd I've never seen that

15:57 yet a large range of objects more users

16:01 for each object more interesting 3d

16:03 backgrounds you know all really kind of

16:08 had the thought of missions if it's so

16:10 well in with bond but not the idea of

16:13 five per level or anything like that

16:16 talking about more and less covert

16:18 missions your bomb storming a train or

16:21 escape from a building with unsilenced

16:23 weapons against stiffer opposition the

16:27 violence will be low-key I say down the

16:29 bottom so I've returned just behavior

16:32 will not be encouraged control you know

16:36 you can't really say too much about this

16:38 this whole paragraph isn't maybe it'll

16:40 be like this maybe it'll be like that we

16:43 didn't have hardware we didn't know

16:46 realism

16:47 I I knew that it shouldn't be totally

16:52 realistic had to feel a little bit

16:54 realistic well but what I had in mind is

16:57 a kind of film realism which is a silly

16:59 realism on the surface it looks real for

17:03 this got bizarre rules you know like

17:06 cars explode the instant you tap against

17:09 someone or you know all the enemies have

17:13 awful aim which is the famous in the

17:18 stormtroopers so I had this kind of idea

17:21 of filming filming realism film physics

17:27 bullets leaving their mark on the walls

17:29 around the level I think that that level

17:32 of interactivity you know pretty much

17:35 all you have is the gun and the more the

17:37 gun can do and reach out to the

17:38 environment and feedback to you that

17:41 stuff is happening the better multiple

17:44 option I was talking about a link out

17:45 between two consoles thank goodness we

17:48 didn't so this is a list of levels which

17:54 is kind of two-thirds how it came out

17:57 we had actually we put down not the

18:00 second level as the first level in the

18:01 end we have seven err are twice as I

18:04 thought that worked out our fav oh I

18:11 find wrote six six levels here at the

18:14 end era they bow so I guess you know I

18:16 wanted there to be a goodly number of

18:18 levels but I didn't feel there was quite

18:21 enough material in the film so trying to

18:23 pull out more and more from the film to

18:27 you know make make a substantial game

18:33 you know here I'm here I'm talking

18:35 through the possibility of missions on

18:38 on each level

18:44 I listed the weapons and looking back

18:51 this is very naive this list because

18:54 it's so ambitious you know I've got a

18:58 sniper's rifle with a site in this list

19:01 and the game the game wasn't supposed to

19:07 take nearly three years to make but you

19:13 can already see the degree of ambition

19:17 within me and also within the team in

19:20 this document various gadgets MQ

19:27 wouldn't ever been great to have

19:29 Moneypenny it was not to be now this is

19:34 an interesting list this is a visual

19:36 effects which I felt was fit to dedicate

19:39 a page to in this short document muzzle

19:42 flashes ricochets cartridge cases

19:45 ejected from guns wounds sparks from

19:50 cute equipment flames smoke day and

19:54 night gas escaping seeing Lake Effect's

19:57 mirror and glass reflections again

20:01 really quite ambitious for what was

20:03 supposed to be a short project so in

20:09 terms of material from the film we have

20:10 very little to begin with we had an

20:11 early shooting script which was lovely

20:16 to thumb through whom ana.k us a lot of

20:19 information about what they were

20:20 planning in the film but we didn't know

20:23 what they were actually going to shoot

20:25 before too long we were informed of

20:28 their shooting schedule and we were

20:31 invited to go along to collect

20:34 photographic material and this is our

20:37 copy of the document they said you can

20:40 see I put a question mark on statue park

20:43 there wondering if it was going to make

20:47 it as a level and if it was worth a

20:49 visit to the film set but all of us were

20:51 very privileged to be offered to go and

20:54 see the film set and

20:56 you see a photograph twitch you know

20:59 perhaps the director Martin Campbell

21:02 might have preferred to have a quieter

21:05 set but he was a great item Norma's

21:07 pleasure knows our norms helped to

21:09 making the game so closely follow the

21:12 film so the next higher to the team was

21:18 Carl Hilton and Carl said to me it was

21:26 great to go down to the sets that leaves

21:27 done and take photos of all the

21:28 different locations

21:29 he also was a Bond fan almost all the

21:32 team we were all Bond fans selection

21:34 effect it was so you know it was a

21:37 project that personally interested all

21:39 of us at the end here he said for me 32

21:43 by 32 RGB 64 by 64 tint the n64 did not

21:52 have a large texture cache and one of

21:55 Carl's bugbears and one of his very

21:57 tight constraints was the size of the

21:59 textures he had to make and most of the

22:03 textures were 32 by 32 many of them

22:06 could be as large as 64 by 64 and now

22:09 you know nowadays I've never actually

22:14 seen but I'm sure there's textures that

22:16 a twelve megabytes which is as large as

22:19 the whole ROM on gold night you know

22:23 nowadays you the smallest iPod you get

22:25 his two gigabytes which is hundred more

22:32 times larger than the entire Goldeneye

22:35 ROM so people have very strict

22:40 constraints to work to this is Carl

22:43 as a scientist being Big Shot for no

22:49 reason and you can see the texture in

22:52 the background there's really not that

22:54 many that many pixels there so Carl was

22:59 good enough to send me some of his heart

23:04 work that was from his desk

23:07 Fiske it's a funny you know how how

23:13 relaxed we were this is his his list of

23:17 what levels were definite he's written

23:26 here again another texture size and

23:28 another one at the top 64 line

23:30 you can tell what was on his mind this

23:34 is his plan for silo version one at the

23:36 bottom here that's the level with halls

23:40 house and was one of the very first

23:42 levels that was modeled we actually

23:45 showed that at Shawshank I on video only

23:48 form with a mock-up it wasn't a playable

23:51 game at all it was a mock-up with

23:53 handcrafted cameras blinds and then all

23:59 the audio was added in afterwards so

24:03 there's a thus a more modern reverse

24:05 engineered image of the silo level

24:12 that's not contemporary just to give you

24:14 an idea this is Karla textures now you

24:33 know as we got through the levels we've

24:35 got a little bit more sophisticated this

24:37 is some petersburg version 3 on the

24:40 right here but really as I as Carl put

24:45 it it's not the art of Prometheus

24:47 you couldn't put this together in a

24:49 coffee table book but this is pretty

24:52 much all the paper that we used during

24:55 the development of the game really what

24:57 we were focusing on was the tools on the

25:00 SGI machines that we had looking at the

25:04 screen they're resizing the window down

25:07 to 320 240 to try and simulate what the

25:13 old 64 would be like

25:18 me Jones was the next hire she was an

25:23 artist and was responsible for many

25:30 things but chiefly or all the characters

25:32 I asked her you know what was the what

25:35 was the work you did that you've most

25:36 proud of and there's a bit of a tease

25:39 but the answer was building the for

25:43 James Bond's each one was different

25:45 Roger has a white tux Shawn had a

25:49 double-breasted jacket and Tim had an

25:52 open tux I was very pleased with the

25:54 likenesses but I guess you'll just have

25:57 to believe me sorry about that this is

26:01 another piece of work that B did which

26:03 was for the promotion of the game this

26:09 was done actually once the game was

26:10 finished and she had a little bit more

26:12 free time first designer on the project

26:15 was Duncan Bach would he he said to me

26:20 that he was most proud of a multiplayer

26:22 mode steve coded it in a month which was

26:27 the work of a magician while I set up

26:30 the spawning weapon and ammo points mrs.

26:33 Duncan talking and cut the level

26:35 backgrounds down to a quarter of their

26:36 poly count then made new backgrounds

26:38 when we didn't have enough variety Carl

26:40 contributed some fantastically complex

26:42 environments as well which meant I had

26:44 to go back and do better texturing on

26:47 the ones I had made you know on a small

26:50 team like this there's so many jobs it's

26:54 impossible to explain what it what each

26:57 person did comprehensively so I can only

27:01 give you a tiny indication in a one-hour

27:03 talk Steve Ellis was a mention

27:08 programmer who did the multiplayer and

27:12 many other things he said one of the

27:14 things that always strikes me as crazy

27:15 in retrospect is until sometime like

27:18 March or April of 1997 so that's two

27:22 years after we'd begun there wasn't a

27:24 multiplayer mode at all hadn't even been

27:26 started it really was put in at the last

27:29 minute something you wouldn't dream of

27:31 in these days and it was done without

27:33 the knowledge or permission of the

27:35 management too rare a Nintendo sorry

27:41 management a rare a Nintendo the first

27:44 thing they knew about it was when we

27:45 showed it to them working however since

27:48 the game was already late by that time

27:50 if we haven't done it that way probably

27:51 would never have happened

27:52 and I suppose there's quite a few

27:55 features that we snuck in and it was

27:59 only possible because of the fact that

28:02 the management trusted us you know they

28:05 believed in the team and you know they

28:07 weren't they weren't coming in even

28:09 every week you know I'm sure there was a

28:12 six-month stretches where no member of

28:16 rare or Nintendo management came in to

28:19 the team offices which is really quite

28:23 extraordinary and all credit to them

28:25 that they felt able to take that much of

28:31 a Leanback approach and place that much

28:35 trust in a team all the people on the

28:39 team had never made a game before except

28:44 for myself and ad Smith so a lot of

28:51 trust down grain nor gate was one of

28:56 three audio people on the project Graham

28:59 was involved for extremely long stretch

29:02 of the project and responsible for all

29:06 the sound effects one of the fascinating

29:08 things about the sounds when you

29:09 actually shoot in Goldeneye and you hit

29:12 something something like seven eight or

29:14 nine

29:14 sound effects will actually trigger so

29:17 much effort was put into the various

29:21 permutations and combinations and the

29:24 randomness in the sound effects to try

29:28 and give a punch and a pleasant tactile

29:32 sensation and you know the reward yes I

29:38 either game no you're hitting this I

29:41 know you're hitting

29:42 I know you're hitting the other so

29:44 enormous amount of effort went into

29:46 polishing compressing you know iterating

29:49 iterating iterating on those sounds and

29:52 you know Graham did all of that but he

29:54 was also a composer and he wrote half of

29:58 them using something like forty or fifty

30:00 pieces of music in the entire game which

30:04 is an incredible feat not least because

30:08 of the measly budget I gave them to fit

30:12 all of the sound effects all of the

30:15 instrument facts and all of the MIDI

30:16 music on the wrong this is the famous

30:24 David dope he was seconded last designer

30:30 to join the team this is him as a

30:34 scientist he was unusual we had a great

30:37 piece of good luck he was actually a

30:40 biochemist his PhD was in biochemistry

30:45 before he before I hired him to rare to

30:50 actually be a system administrator for

30:52 rare so after a while I'm talking to him

30:55 you know we had a great relationship and

30:58 I felt that he had been an awesome

31:01 addition to the team I'm so glad I made

31:05 that choice um who's you unquestionably

31:09 one of the best FES designers in the

31:13 world

31:13 this is Dave dhoka myself being silly

31:17 you know the team had to produce you

31:21 know we we scratched for what

31:24 photographic materials we could get but

31:27 there's very little in the way of faces

31:29 of the actors even even the bit parts

31:35 even even pierce brosnan so it was like

31:38 magazines and we found a picture of jaws

31:40 it was about this big in some magazine

31:43 so yeah of course crazy

31:45 here we are we're dressing out to

31:47 provide silhouettes for the video

31:50 screens in 7 a.m. bunker

31:53 this is actually a later picture from

31:56 near the end what I've got in my hand

31:57 there is the mock-up of the box art for

32:02 the product that was produced by

32:04 Nintendo they sent it to us we printed

32:06 it out on one of the few color printers

32:08 in the company and I folded it together

32:11 to make a simulation of what the box

32:14 would look like one of the things I'd

32:17 like to talk about in Goldeneye that I

32:20 think isn't spoken about enough is humor

32:25 there's a lot of things that happen in

32:28 there some were intentional some not so

32:31 intentional but I think all of them

32:33 great addition to the game there helped

32:36 to make you laugh any any first-person

32:39 shooter and almost any game is a series

32:45 of hills and valleys of tension you know

32:49 you go around the corner

32:50 maybe there's a guy there oh there isn't

32:53 you go into another corner no guy oh I

32:58 hear some shooting over there come

33:00 around a corner three guys bang bang so

33:02 it's that's ramping up the tension but

33:06 to bring it down again it's more

33:08 difficult and humans incredibly

33:10 important for that and you know that

33:13 helps to bring the valleys in and to

33:15 make the to give the level of structure

33:17 and emotional structure which is really

33:22 important for people's people's reward

33:24 because people are base emotional

33:26 animals so you know if it's not really

33:35 realistic it was it was kind of cool

33:38 maja mark was very pleased the first

33:43 time he got the watch to come up on the

33:45 screen but you know really it's it's a

33:48 little bit silly and the explosions

33:54 there's another example of humor because

33:57 for reasons I still don't understand

34:00 when there's an explosion it's kind of a

34:04 crediting effect and the software

34:06 credits the momentum to the nearby

34:09 enemies and when it reaches a certain

34:12 amount they fly off so what you happen

34:15 is there's an explosion people are there

34:17 like this slightly too long and then

34:20 they fly off into the air and it is very

34:24 very funny and you know I by rights I

34:28 should have been sick of the game by the

34:30 end of 2 and 2/3 years but I derived a

34:36 great deal of satisfaction from just you

34:38 know lobbying around a few few minds

34:41 blowing out people watching them cascade

34:44 around you know the silly there's an

34:49 argument that we could have had more

34:50 quality control and we can say yeah you

34:52 know what

34:53 let's not have the Moonraker outfits

34:55 they're a bit silly but instead you know

35:00 be made it and it went in this it is

35:05 silly but it's also great yeah thus

35:07 laughing oh how did nobody ever said on

35:12 the team no one ever said this this is

35:15 ridiculous it was one of a very very

35:20 long list of things you know we wanted

35:22 it we implemented it

35:24 it seemed okay we move on you know we

35:27 it's it's absolutely hilarious and I

35:29 love it but again you could make an

35:33 argument that it should be different

35:35 more realistic much more sober and I

35:38 think you'd be taking out a lot of

35:41 character of the game the toilets yeah

35:44 the English and Japanese found humor in

35:47 the talk so you know overall you know

35:53 what I what I love to try and put into a

35:55 game is a sense of humor you know I

35:58 think people want to have fun and a part

36:01 of that is laughter but also any game

36:03 it's presents a challenge it can become

36:05 become tiresome the frustration can

36:07 become overbearing the more humor you

36:09 can get in there the better and a lot of

36:12 it is going to be serendipity it's going

36:15 to be things you do that

36:17 you didn't plan on what they turn out to

36:18 be a little bit silly and if you can if

36:20 you can leave those and the teams

36:23 comfortable to leave them that's correct

36:27 next thing I want to talk about is

36:29 mistakes I like things that by rights

36:32 are really truly a mistake in game

36:35 design I guess some of you might know

36:41 I'm probably the biggest one is odd-job

36:46 have really anxiety in you know odd job

36:55 was an annoying character in multiplayer

36:57 because it's very hard to shoot because

36:59 he was so short and you could certainly

37:03 make an irony should never have been in

37:05 the game that it be a better game

37:08 without him some you know one of your

37:10 friends is going to choose all job it's

37:12 a cheesy thing to do and it spoils the

37:14 game but to me that's a social dynamics

37:18 problem you should be putting pressure

37:21 on your friend to not be a cheater but

37:24 fearful of people have redefined this as

37:26 you know our job is what the cheater

37:28 picks and I think that's a beautiful

37:31 solution to what could be said it was

37:33 under balancing we didn't have time to

37:39 bounce all the characters really we just

37:41 threw them in for the multiplayer but

37:45 you know these little these little

37:47 quirks and bits of sketchiness I really

37:51 strongly feel I bring personality to the

37:53 game another one of the things that I

37:58 personally feel so proud about the game

38:00 is kind of sandbox nature it's not it's

38:04 not really it's not really a sandbox

38:07 game but many of the levels are linear

38:12 and many of them you you you have a

38:15 little bit of choice you know you could

38:17 do the missions in this order all that

38:19 order but really you know is you can't

38:22 really say it's the sandbox game but

38:23 there's there's a there's a sandbox a

38:26 way to play it and that's when you don't

38:29 care about the missions

38:30 anymore when he played it enough and you

38:33 just want to mess around with with the

38:36 guys and you know my personal fave I

38:40 spent I remember well six hours one

38:44 evening playing and playing in in the

38:48 bunker drawing in more of the infinite

38:51 forn

38:52 enemies and shooting their hats off six

38:57 hours and I had I had like a massive

38:58 collection of hats on the ground and you

39:01 could shoot them I don't know if you

39:02 ever know this you can shoot them you

39:04 can get them up in the air you can

39:05 juggle the hats you can chew them at

39:08 Calculon party you can see you know that

39:13 that is a really pleasurable thing and

39:16 it's something that came out of the game

39:18 development totally organically nobody

39:20 ever said you know let's make it so they

39:24 you know so that you can shoot the house

39:26 in the air for six hours it would have

39:30 been a ridiculous statement but you know

39:32 what we did say was that's a kind of

39:35 funny why don't we have some hats and

39:39 the rest was mark Edmonds execution now

39:47 one stage we did have some much more

39:50 violent effects in the game we have one

39:52 texture long animated texture new masses

39:56 of blood so when you shoot it was made

39:59 me feel you know I consider myself to

40:01 have a strong stomach made me feel a

40:02 little bit uncomfortable seeing this in

40:05 glorious 3d you know Howard Lincoln came

40:09 round one time and he was he saw one of

40:12 our animations which was to shoot

40:14 someone in the groin and they then would

40:17 ride around on the ground for it seemed

40:20 like it must have been less than two

40:22 minutes but seemed like forever that

40:24 they were there writhing around on the

40:26 ground and it had a humor value but

40:30 really it was too too violent you know

40:33 and we'd we took out the the flip book

40:36 texture or incredibly bloody we toned

40:39 down the hits so there

40:40 really quite subtle on the bodies that

40:44 you know they are read the characters

40:47 disappear after you have shot them

40:50 totally not realistic but it kind of

40:52 underlines the fact that you shouldn't

40:56 really care about them that you know you

40:59 shouldn't feel bad about shooting them

41:01 it makes it easier to to play a game

41:03 which is you know like any FPS it is

41:06 essentially about killing a long long

41:08 series of Fievel and I have to say the

41:12 graphics which today looks so blocky and

41:15 awkward you know even in the day they

41:17 were a little you know the shape of the

41:19 head it wasn't it's a little bit you

41:22 know can't we do better than this

41:24 but that non-realism in the graphics was

41:29 actually a big help to making it a more

41:32 light game and more in touch with the

41:36 Bond universe so grant kirkhope was the

41:43 other main composer on the project he's

41:49 also a guitarist and he did the guitar

41:52 riffs on the some of the more electric

41:56 rock versions of the gold my Monte

42:00 Norman piece grant said to me I wish I

42:03 had added longer tunes for the game I

42:05 remember asking mr. Hollis how long I

42:07 should make the chains as I was

42:09 completely new to games and I had no

42:10 idea I remember him saying that he

42:13 thought a minute would be sufficient and

42:15 quickly went back to his business and

42:16 dismissed me I'll try to know that you

42:20 could spend forever running around

42:21 levels hearing the same tune over and

42:23 over and over again thanks Martin sorry

42:27 sorry gran Robin beans and was the third

42:31 to compose the the music guys were the

42:35 only people who are out of the team

42:37 building you know the the audio team was

42:40 all together and rare was privileged to

42:42 have these three guys who were

42:44 incredibly untalented musicians

42:49 composers from the professional

42:53 world rather than games design plus what

42:57 rather than game development per se Ron

43:00 Wieland

43:00 only wrote one piece of music which was

43:03 the elevator music which you can hear in

43:05 such levels as the the complex and the

43:09 caves this is just a little joke but I'm

43:12 very interested to Adrian Smith was the

43:17 third and last artist he said to me I

43:20 was particularly pleased with the

43:22 effects I produce for the game

43:24 the muzzle flashes the trace of fire in

43:26 the explosions we're very influenced

43:28 with the film heat I was very pleased to

43:31 the end sequence of the blood coming

43:32 down the screen I wish I had copyrighted

43:34 the watch graphics as a mechanic for the

43:36 status for health indicator mentioned

43:46 Stephen Ellis so here's a his a nice

43:48 shot of the multiplayer this is actually

43:52 running on an emulator if you blow up

43:55 the n64 screen on a big screen like this

44:00 you know ntsc was the major limiting

44:03 factor it's really not in the in the

44:10 actual bandwidth for ntsc but you know

44:13 this is one of the things that I feel is

44:15 held at least well about the game is is

44:18 the sharpness I asked Adrian what he

44:23 what he wish had he put in to the game

44:26 and he said the firing range this is

44:28 something he actually modeled it was an

44:31 environment from q-branch it would have

44:32 been great if it was added and

44:33 incorporated with some minigames well

44:37 that was one of the very few things that

44:38 we didn't get around to actually getting

44:40 in but it came true with perfect dark so

44:45 you know by this point we had the entire

44:48 team assembled and you we're in the

44:51 homestretch I I have to admit that I

44:54 don't remember any talk of deadlines or

44:59 schedules which is completely astounding

45:04 you know I assume that as lead on the

45:09 project I was told what the deadline was

45:12 probably several times but I was so

45:18 focused on what what could be done to

45:23 make the game better that you know today

45:26 this isn't in my mind at all

45:29 now this is actually a graph by Ben

45:33 cousins that he sent round on Twitter

45:36 recently to try and educate people who

45:39 don't know game development how quickly

45:42 your game will actually develop what you

45:44 think will happen is you think linearly

45:46 your game will get better half way

45:48 through you'll think you'll have half a

45:49 game it's not the case at all halfway

45:51 through you basically don't have

45:52 anything you don't have a game so

45:56 halfway through goldeneye you know we

45:58 did have a game and i remember sitting

46:00 in the office with mark Edmunds and he

46:02 said you know come and have a go I had a

46:04 go and I just remember not really

46:09 feeling anything and I didn't say

46:11 anything and I just I guess my faith

46:15 said everything yeah it was it was

46:17 absolutely diabolical it was in the

46:19 bunker in the interior the movement was

46:22 completely dead the sound effects are

46:25 completely dead the gun was dead on the

46:27 screen and it had no no reaction no

46:29 movement there was no auto aim it was a

46:32 crosshair you know you had to line up

46:34 perfectly was completely no no juice at

46:37 all in the game and you know essentially

46:42 it took circa a year to add a layer on

46:46 layer on layer of little movements

46:48 little sounds little reactive effects

46:52 graphical effects on textures when you

46:56 shoot little glimpse on the gun you

47:01 layer upon layer upon layer of little

47:03 effects just to try and make it feel

47:04 more more punchy and more like the game

47:08 was actually listening to what you were

47:11 doing

47:13 and in this final year we were working

47:16 so hard I I can't remember it at all

47:20 so I have almost met what to say about

47:25 that new the homestretch

47:27 the game was finished we were able to

47:30 add little little sprinkles little

47:34 additions things that were just nice to

47:36 have like the like the cheats you know

47:41 Steve Ellis programmed these we just had

47:44 a great discussion in the office all you

47:45 know what yeah that would be cool yeah

47:47 why don't we have big heads like Donkey

47:49 Kong Country you know that'll be funny

47:52 tiny bombs yeah that's amusing and you

47:56 know we added it gathered cheese and

47:59 just I think we had seventeen it was

48:00 just the number we could think of that

48:03 we thought would call a little bit of an

48:07 intro and an outro on the game this is a

48:10 joke referencing the BBFC their film

48:15 review board in the UK and this is my

48:19 signature at the bottom left and my

48:22 attempt to do to write James Bond how he

48:25 would sign it there yeah I mean all this

48:31 was added right at the end at this point

48:33 we finally had a full complement of

48:36 levels 18 from the Goldeneye movie and

48:40 two extra bonus from other films you

48:47 know a little story about the the

48:51 balancing on that you Dave doke and I

48:56 played the Aztec after he had set it up

48:59 and we both found it very hard on

49:04 double-o but you've given a couple of

49:09 hours we could both we could both do it

49:13 and then being a mischievous pair that

49:16 we were I said to him okay that's good

49:19 now why don't you just make it a little

49:22 bit harder so a snake is just slightly

49:26 harder than we were able to do ourselves

49:30 at the time and it's in my if I give Li

49:34 the hardest level in the game on

49:36 double-o but I still get a great deal of

49:38 satisfaction from playing Hilter I can't

49:40 be frustrating at times so Mark had a

49:45 little story points out to me that I've

49:47 forgotten about which was the craziest

49:50 coolest thing we finished the game it

49:53 had been through tests

49:54 you know we'd marked off all the bugs

49:55 everything was good

49:57 they'd send it through lot check words

49:58 there's a long process where they play

50:01 the game to death for like two weeks and

50:05 tested on all the combinations of

50:07 televisions and so on

50:08 and then he came back and there was a

50:10 problem on one of the levels when he

50:13 played it in a certain order it's

50:15 probably the frigate feeling so an order

50:18 very certain sequence of characters

50:20 being instantiated and they'd have

50:22 terrible texture maps on them because

50:24 there wasn't enough memory had a dynamic

50:26 memory system which is I mean it worked

50:30 okay most of the time but memory is very

50:32 tight on this level so what I had to do

50:35 was right in order to make a minimum

50:38 change and the version control system

50:40 wasn't that reliable in order to make a

50:42 minimum change that day I wrote at all

50:46 to extract the code and the data for the

50:52 code from the ROM it was gzipped

50:55 so basically a custom on Jie Jie I wrote

51:00 that adjusted the memory numbers rajee's

51:03 appear whopped it back into the ROM

51:07 image without recompiling everything

51:10 anything and sent it to a Nintendo

51:15 so that's the final version and mark

51:19 says thankfully it worked and it still

51:21 makes me smile it's kind of I don't know

51:26 I don't know if it was the best way to

51:28 do that or not to this day but it was it

51:32 was a minimum change and it worked

51:39 another little story from the final days

51:42 which was the crates used to say vodka

51:45 on them which is Russian for vodka

51:48 unfortunately pretty late on Nintendo

51:51 noticed they have who spoke Russian and

51:55 said yeah but mine large almost all the

52:00 things we wanted to get into the game

52:02 were we were able to fit them in with no

52:06 resistance from rare management or

52:10 Nintendo or even eon and Don Jack who

52:15 are the production company that made the

52:17 movies and the licensed controlling

52:20 company so you know I'd like to say

52:23 thank you to all of those people for

52:25 allowing us to realize our vision and

52:30 then the game was finished and I guess

52:33 it was a tiny bit of an anti-climax but

52:35 there was a certain amount of pleasure

52:37 you know being able to being able to

52:40 look at things that nearly three years

52:42 before had just been a few words you

52:45 saying the dam would have three towers

52:47 and now you know being able to enjoy

52:50 mm-hmm enjoy our work mark Edmund says I

52:57 remember Tim asking shortly after we had

52:59 finished what I would raid the game I

53:01 think I said seven out of ten so there

53:04 must have been a lot of areas I still

53:06 thought could have been improved even

53:08 back then and I think that sums up how

53:12 all of us felt about about the game we

53:16 didn't really know whether it was

53:19 awesome or whether it was okay or

53:21 whether it would

53:23 you know fall on deaf ears in the

53:25 marketplace because the film had been a

53:26 long time ago and you know to our eyes

53:29 the graphics on the levels we've made

53:31 first were not as good as the graphics

53:33 on the levels we'd made last so there's

53:35 a lot of questions in our minds this is

53:37 the closest thing I've got to a gold my

53:39 team photos is not the gold my team this

53:41 is part of the perfect dark team from

53:44 left to right we've got Steve Alice

53:45 Graham nor gate mark Edmonds that's Ross

53:49 Murray who was perfect dark only Karl

53:52 Hilton David dog myself Pete Jones and

53:56 Lee ray who was also perfect dark only

54:06 so Oh - oh no - in my opinion is that

54:13 it's really quite toxic to go into

54:18 making a game with the idea that one

54:20 person is going to be giving all the

54:22 orders and telling everybody what to do

54:25 I think that's absolutely the worst way

54:27 to go about making a game and I tried

54:31 yeah and I always try but I tried on

54:33 this project to as much as possible have

54:37 a flat management structure and have it

54:41 so the people could implement their own

54:42 feed shows if they felt like it and

54:45 people could similarly contribute

54:47 whatever it was that they wanted my

54:50 chief concern in the early days was to

54:55 simply direct a ship in what seemed like

54:58 a good direction and make sure I had the

55:01 best possible crew and the thing is when

55:04 you work with incredibly intelligent

55:06 people they they reward you know your

55:12 trust in them and the most rewarding

55:15 thing for them is their autonomy and

55:20 that means not telling them what to do

55:23 so really you know my take home for this

55:28 talk is

55:30 les you can be a boss the less you can

55:34 be telling everybody what to do

55:35 the better the likely results auteur no

55:41 tour I think it's a false dichotomy

55:43 really I think the best games you do

55:47 have an element of personal an element

55:50 of personal authorship but you really

55:54 need that very early in the project

55:56 before the team is really a team at all

55:59 because if you're trying to steer a team

56:02 around and they're great people they're

56:05 going to resist you and it's going to

56:07 have an emotional cost on both sides so

56:13 really I think the best thing is just

56:15 just a couple of drops of yourself in a

56:18 project and after that everybody

56:23 collaborating Adrian Smith said it best

56:28 to me he said everybody contributed to

56:31 the game in their own way which made it

56:34 the success that it became nobody needs

56:37 to be singled out it was a team that

56:40 made gold Knight just like say thank you

56:44 for the gold my team with their help

56:46 with this talk

56:48 [Applause]

57:02 I've also been asked to remind you to

57:05 fill out your mobile evaluations thank

57:09 you very much I'm happy to take

57:12 questions although I talked for what's

57:17 also lunchtime at the 4th floor so

57:20 questions it's fine too okay a lot of

57:27 people fight about first-person shooters

57:30 on consoles versus PC like mouse

57:33 keyboard and all that stuff how do you

57:35 feel that even now so many years after

57:37 Goldeneye came out when people talk

57:39 about whether or not shooters can work

57:41 on consoles they still mentioned

57:43 Goldeneye oh they do that's cool um how

57:49 do I feel that people still mention

57:51 Goldeneye when debating whether

57:55 first-person shooters can work on

57:57 console versus PC I mean to me it seems

58:03 a little bit sad some of the acrimony

58:05 between the two camps the you know PC is

58:08 the one true way camp and the console

58:12 camp I think I think they're both you

58:14 know you can make awesome awesome games

58:16 on either you know people will say well

58:19 the analog the little analog stick it's

58:21 not so accurate it's not as accurate as

58:24 a mouse but to me it's like saying in

58:26 golf why do you just pick up the ball

58:28 and put it in the hole you know I mean

58:33 the game comes every game comes with its

58:36 own constraint and playing the game you

58:40 you live inside those constraints and

58:42 they they make the game so it's kind of

58:46 a strange thing to say to me PC is

58:50 designed for spreadsheet software it's

58:53 incredible really that you can make such

58:55 an awesome game on it only other hand

58:57 consoles they have the price pressure

58:59 but at least they're designed dedicated

59:03 for games you know I never felt any need

59:05 for a divide between PCs and consoles

59:08 I never thought oh we can't make a great

59:10 game

59:11 because doom or no or anything like that

59:14 I was always thinking well you know

59:16 let's let's go for this I think it's

59:20 also if people still mention gold night

59:21 I mean at the time it was the

59:23 best-selling FPS that's a great

59:27 privilege and honor we've had a huge

59:30 amount of work into making it as crazy

59:32 could be in us it's lovely to be invited

59:35 back to to talks you know is pretty much

59:38 15 years slightly over 15 years since it

59:41 came out

59:55 did you play the well let's call it

59:58 remake that came out recently and what

60:00 do you think about the new iteration of

60:03 Goldeneye

60:05 I played eurocom Activision's remake a

60:10 little bit honestly not that much I

60:14 thought I thought that it is really

60:20 close to a Call of Duty more in its

60:24 gameplay and the rhythm of the enemies

60:29 the rhythm with which they come to you

60:32 the way you approach them you know and

60:37 and in many other ways although

60:38 graphically I could see they put a huge

60:40 amount of work into trying to be true to

60:43 Goldeneye double-oh-seven on the n64 you

60:48 know I think it's it's actually very

60:50 difficult to analyze a game it's

60:53 actually very difficult to see what's in

60:54 there I think it's actually very

60:55 difficult to replicate that so yeah I

60:59 enjoyed playing it maybe they were

61:03 trying to cater to the marketplace today

61:06 more and very different I mean if that

61:08 was my job I I don't know that I would

61:12 say no make it just like the original

61:14 yeah it's very difficult because there

61:17 aren't really games in the marketplace

61:18 today that have that plate-spinning

61:19 thing where you you have like three

61:22 guards and you shoot one that keeps him

61:24 busy

61:24 shoot the next keeps him busy shoot the

61:26 next keeps him busy then you go back to

61:28 the first one again you know that

61:30 there's lots of little mechanics like

61:32 that in gold Knight to do with the

61:34 second to second and quarter second two

61:37 core second interactions between you and

61:39 the guards yeah even stuff like the

61:43 bunny hop they do it's so silly but it

61:46 makes a big difference to the business

61:49 of actually trying to shoot them you

61:51 know so you're you hamstrung today by

61:54 realism and by the games they're

61:56 successful in the marketplace so you

61:59 know perhaps that's why they were

61:59 thinking I don't know

62:15 it sounded in your talk as if the first

62:18 time you guys figured out that it was in

62:20 any way successful was when they called

62:23 you jaws to give this talk so so what

62:28 was like when did you guys start to feel

62:31 like hey this game kind of matters

62:33 someway yeah when did we

62:36 when did the team realized hey this game

62:39 kind of matters in a way you know I

62:45 remember going to e3 very near the end

62:47 I'm feeling extremely pained other

62:50 things some people said when they came

62:52 up there were there were eight consoles

62:54 with gold nine people who come up and

62:56 playing some people said oh look

63:00 graphics are amazing and then it's like

63:02 yes score and next person come like

63:05 water peace so yeah not then that was

63:18 that was 95% of the way through

63:20 developing the game we had like five six

63:22 seven levels in working we were

63:25 basically nearly done we shouldn't have

63:27 gone to e3 we should have been back in

63:29 twycross zoo today didn't sell a massive

63:38 spike so it was really debatable if us

63:43 at success once it was launched and then

63:46 it came out in Japan and much of Europe

63:54 David Oak has a graph of the rentals for

63:58 goldeneye in the US is very interesting

64:01 it's completely flat graph it just goes

64:04 on for about two years it just goes on

64:07 and on and on the game just sold you

64:10 know not millions it sold like a hundred

64:12 thousand two hundred thousand a month

64:14 and then again and then again and again

64:17 and that was I guess when we realized it

64:21 was it was something it was going to

64:23 sell more than a million copies it's

64:27 going to sell more than two million you

64:29 know gradually we realized it was going

64:32 to have commercial success but but

64:36 really you know I still don't know and I

64:40 warned her that people still going to

64:44 mention it in another 15 years I think

64:46 he's still going to mention it in

64:47 another 50 years so you know games can

64:52 come and go they can be talked about for

64:54 a while so you know really I still don't

64:56 know whether it will be eternal yeah

65:06 that's true yeah well you know the

65:10 golden team was all spread around now a

65:13 couple of guys in Canada in the US and

65:17 spread around the UK but still yeah

65:19 that's a good idea

65:36 hey Golan I seem to be this kind of

65:39 perfect storm of like eight guys has

65:42 just never made a game before your

65:46 management seem to be just yeah go ahead

65:47 do what you like there's kind of really

65:50 free happy time of just game development

65:52 that has become the quintessential bond

65:55 game every game since has tried to

65:57 recapture that spirit and that

65:59 excitement do you think that's ever

66:02 doable again now that the game industry

66:04 has gotten so serious do you think

66:05 that's it was a one-time thing of

66:07 everything coming together and making

66:09 this great game so Goldeneye was a

66:15 perfect storm of everything coming

66:19 together you say and it's the

66:24 quintessential bond game that people try

66:27 and live up to is there any possibility

66:29 that that could react our first of all I

66:33 have no subject I have no objectivity

66:36 you know I I'm completely in love with

66:40 my games every one of them and I don't

66:45 feel you know I should be raised up on a

66:50 podium in a position of authority to

66:52 decide that I know that I personally

66:55 love Goldeneye in the way it turned out

66:58 but you know whether whether it will

67:03 always be the best bond game I really

67:08 cannot say you mentioned one thing I'd

67:11 like to pick up which was I gave the

67:13 impression you know it was all it was

67:16 all plain sailing and management was

67:17 like yeah do whatever you I do have some

67:19 stories of things things they wanted to

67:21 happen the you know mainly didn't but

67:26 occasionally did happen

67:27 wish I'd be happy to tell if anybody

67:30 wants to hear them but 99% of the time

67:34 yes it was we got to do what we wanted

67:51 that seems to be it for questions thank

67:53 you ever so much everyone

67:55 [Applause]

67:56 [Music]